The Alan Turing Institute Programme

September 2018 to August 2024

Human-AI Interfaces and Robotics

Programme Director: Professor Sethu Vijayakumar FRSE, University of Edinburgh

Creating an economy that harnesses artificial intelligence (AI) and big data is one of the great opportunities of our age. Innovative scientific developments in the field of Artificial Intelligence must continue to play an important and growing role in meeting this ambition – with a potential contribution of £200 billion or 10% of UK GDP by 2030.

On a closely related front, the market impact from Robotics and Autonomous Systems (RAS) innovations is estimated to be £1.6-£6.4 trillion per annum by 2025. Smart, connected machines, or robotics and autonomous systems (RAS), are acting as tools to support us, working alongside us, making independent decisions and even learning from experience. They act and sense in the real world, connected and collaborating through the internet of things, generating and enabled by large quantities of data, using artificial intelligence to reason, classify, control and interact. They are emerging from research prototypes into practical applications.

Cross sector RAS exemplars that are already emerging and in place include; autonomous and semi-autonomous cars; manufacturing systems that can personalise bespoke designs and reconfigure; robotic fulfilment centres that assemble, package and dispatch goods ordered online; drones that deliver packages, or map, inspect and repair in our offshore oilfields and nuclear facilities; assistive exoskeletons to help us move and lift; surgical systems for the most intricate of operations; and interactive companions for the elderly and isolated.

There are natural synergies to be exploited that will bring together the unique strengths of UK research in machine learning, data science and the recently established cutting edge programmes in Robotics and Autonomous Systems to develop a distinctive and world leading programme.

Programme Vision statement

To create a world leading scientific programme of data driven AI research and innovation that addresses the unique challenges arising from and towards deployment of Robotics and Autonomous Systems (RAS) technology for solving socially relevant problems across domains.

Aims and Objectives

The key aim of the programme is to develop and support a world leading portfolio of activities that will lie at the intersection of data driven AI and machine learning, specifically targeted to the RAS domain. This will be achieved through:

building and funding a core research team investigating ‘fundamental’ algorithmic and computational innovations under three key strands (see below), with a research team lead heading each strand.
develop Joint Industry Projects (JIPs) through deep dive engagements with industry for technologies that are medium to high TRL levels – with between 50-75% of core funding coming from industry.
de-risk deployment through proof of concept implementations in several partner ‘living labs’ that integrate hardware and data processing challenges

Research Agenda

The core research agenda will be focused through three strands [that are crucial but are missing or underrepresented elements in the current RAS and machine learning roadmap]:

Scalable algorithms under constraints
1. Real time inference requirements, computational constraints of embedded, untethered and mobile platforms, hardware limits (torque, joint) for guaranteeing safety, approximate hierarchical inference for graceful performance degradation
Methods for efficient multiagent computations
1. Intention detection and movement prediction, scalable multiagent adversarial and collaborative policies, multimodal sensor aggregation for decision making
Verifiable, Robust and Explainable decision making for multimodal RAS assets
1. Enabling secure systems – communication, decision making; understanding and predicting failure modes, developing robustness and multiple failure recovery modes, fault and risk inference through probabilistic modelling

Each of these strands of research will be closely grounded in the RAS context and will need to have a grand challenge counterpart that it will/can enable, if successful. However, we will actively look for synergies where some of the results (for example in explainable and verifiable AI) can be applied to domains beyond RAS such as IoT, medical diagnostics.

Below are period reports in reverse chronological order.

Programme Updates April 2024 to September 2024

Highlights

Michael Mosley, Valkyrie and Sethu Vijayakumar

Sir Hugh Cairns Lecture at the Society of British Neurological Surgeons (SBNS 2024), 19 April 2024, Edinburgh International Conference Centre [news item]
UoE-DUK Partnership for CoE in AI announced at India's first GenAI Conclave in Kochi, 11-12 July 2024 [news item]
Humanoid and Exoskeleton research featured on Channel 5 Documentary Michael Mosley: Wonders of the Human Body, 29 August 2024 [news item]
Keynote at Wold Robotics Conference 2024, 21-25 August 2024, Beijing, China [news item]
AI in Healthcare Workshop, 30 May 2024, Edinburgh [news item]
ICRA 2024 and ATR Symposium, 13-17 & 20 May 2024, Japan [news item]

Papers submitted or published

Lei Yan, Theodoros Stouraitis, Joao Moura, Wenfu Xu, Michael Gienger, and Sethu Vijayakumar, Impact-Aware Bimanual Catching of Large-Momentum Objects, IEEE Transactions on Robotics (T-RO), vol. 40, pp. 2543-2563 (2024) (Presented at: 40th Anniversary of IEEE International Conference on Robotics and Automation (ICRA@40), Rotterdam, Netherlands) [pdf] [DOI] [video] [citation]

Namiko Saito, Joao Moura and Sethu Vijayakumar, Long-horizon Manipulation through Hierarchical Motion Planning with Subgoal Prediction, Proc. 40th Anniversary of the IEEE Conference on Robotics and Automation (ICRA@40), Rotterdam, Netherlands (2024). [video] [citation]

Joao Moura, Theodoros Stouraitis, Namiko Saito and Sethu Vijayakumar, Optimal Shared Autonomy for Contact-rich Robotic Manipulation, Proc. 40th Anniversary of the IEEE Conference on Robotics and Automation (ICRA@40), Rotterdam, Netherlands (2024). [citation]

Namiko Saito, Joao Moura, Hiroki Uchida and Sethu Vijayakumar, Latent Object Characteristics Recognition with Visual to Haptic-Audio Cross-modal Transfer Learning , Proc. 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024), Abu Dhabi, UAE (2024). [video] [citation]

Andreas Sochopoulos, Michael Gienger and Sethu Vijayakumar, Learning Deep Dynamical Systems using Stable Neural ODEs, Proc. 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024), Abu Dhabi, UAE (2024). [citation]

Juan Del Aguila Ferrandis, Joao Moura and Sethu Vijayakumar, Learning Visuotactile Estimation and Control for Non-prehensile Manipulation under Occlusions, Proc. The Conference on Robot Learning (CoRL 2024), Munich, Germany (2024).

Victor Leve, Joao Moura, Namiko Saito, Steve Tonneau and Sethu Vijayakumar, Explicit Contact Optimization in Whole-Body Contact-Rich Manipulation, Proc. IEEE-RAS 23rd International Conference on Humanoid Robots (Humanoids 2024), Nancy, France (2024).

Jiayi Wang, Saeid Samadi, Hefan Wang, Pierre Fernbach, Olivier Stasse, Sethu Vijayakumar and Steve Tonneau, NAS: N-step computation of All Solutions to the footstep planning problem, Proc. IEEE-RAS 23rd International Conference on Humanoid Robots (Humanoids 2024), Nancy, France (2024).

Funding: Further funding, leveraged funding/support, in-kind contributions

Japan Science and Technology Agency (Moonshot Goal 3) Hierarchical Motion Planning Framework for Realising Long-horizon Task

Project Duration: April 2024 - November 2025 [Project Page]

Japan Science and Technology Agency (Moonshot Goal 3) Monitoring and Intervention in Neural Degeneration

Project Duration: June 2024 - November 2025 [Project Page]

Programme Updates October 2023 to March 2024

Highlights

AIUK Fringe – Innovative Robotics for Tomorrow’s Assisted Living Open lab event, Friday 29^th March
Professor Vijayakumar invited speaker at British Council GREAT Talks, Dhaka 11 February
Professor Vijayakumar opening keynote at AI Conference, Nepal 3 March
Lab show and tell for Regional First Lego League contestants, 23 February

Upcoming plans/priorities

Cooking Robotics workshop at ICRA 2024
Edinburgh Science Festival 2024 – meet the robots event
AI in Healthcare Workshop – Edinburgh, 30 May 2024
Follow up workshop for UK-Italy collaboration to be confirmed

Conference/workshop talks presented

Professor Vijayakumar gave the opening keynote at AI Conference in Nepal on 3 March 2024 [News item]
Professor Vijayakumar was an invited speaker at British Council GREAT talks in Dhaka on 11 February [News item]

Papers submitted or published

Russell Buchanan*, Adrian Röfer, Joao Moura, Abhinav Valada and Sethu Vijayakumar, Online Estimation of Articulated Objects with Factor Graphs using Vision and Proprioceptive Sensing, Proc. IEEE International Conference on Robotics and Automation (ICRA 2024), Yokohama, Japan (2024).
Andreas Christou, Antonio J. del-Ama, Juan C. Moreno and Sethu Vijayakumar, Adaptive Control for Triadic Human-Robot-FES Collaboration in Gait Rehabilitation: A Pilot Study, Proc. IEEE International Conference on Robotics and Automation (ICRA 2024), Yokohama, Japan (2024)
Marina Y. Aoyama, Joao Moura, Namiko Saito and Sethu Vijayakumar, Few-Shot Learning of Force-Based Motions From Demonstration Through Pre-training of Haptic Representation, Proc. IEEE International Conference on Robotics and Automation (ICRA 2024), Yokohama, Japan (2024)
Keyhan Kouhkiloui Babarahmati, Mohammadreza Kasaei, Michael Mistry and Sethu Vijayakumar, Robust and Dexterous Dual-arm Tele-Cooperation using Adaptable Impedance Control, Proc. IEEE International Conference on Robotics and Automation (ICRA 2024), Yokohama, Japan (2024).
Jiayi Wang, Sanghyun Kim, Teguh Santoso Lembono, Wenqian Du, Jaehyun Shim, Saeid Samadi, Ke Wang, Vladimir Ivan, Sylvain Calinon, Sethu Vijayakumar, and Steve Tonneau, Online Multi-Contact Receding Horizon Planning via Value Function Approximation, IEEE Transactions on Robotics (T-RO), vol. xx(yy), pp. xxxx-yyyy (2024) (Presented at: 40th Anniversary of IEEE International Conference on Robotics and Automation (ICRA@40), Rotterdam, Netherlands
Carlo Tiseo, Wolfgang Merkt, Wouter Wolfslag, Sethu Vijayakumar and Michael Mistry, Safe and compliant control of redundant robots using superimposition of passive task-space controllers, Nonlinear Dynamics, vol. 112, pp. 1023-1038 (2023)
Daniel Gordon, Andreas Christou, Theodoros Stouraitis, Michael Gienger and Sethu Vijayakumar, Adaptive Assistive Robotics: A Framework for Triadic Collaboration Between Humans and Robots, Royal Society Open Science, vol. 10: 221617 (2023).

Programme Updates April 2023 to September 2023

Highlights

Edinburgh Science Festival Lab Open day showcased our robot platforms and looked at how robots can shape the future, 14 April 2023
Professor Sethu Vijayakumar delivered one of the prestigious keynotes at ICRA 2023, titled From Automation to Autonomy: Machine Learning for Next-generation Robotics on 1^st June 2023
ICRA 2023 full day workshop, bringing together world leading researchers from academia and industry focusing on contact-rich manipulation scenarios – 2 June 2023
Joint Edinburgh Centre for Robotics and Alan Turing workshop/symposium with the Japanese Moonshot Initiative – 5^th June 2023

Upcoming plans/priorities

Live demonstrations at Kawada, Japan (Nov 2023).
Follow up workshop for UK-Italy collaboration to be confirmed.

General programme updates

Past Events

Edinburgh Centre for Robotics / Alan Turing workshop/symposium 5^th June 2023

Title: Bridging International Robotics and AI Grand Challenges --Working with the Japanese Moonshot Initiative

The workshop will include Professor Toshio Fukuda, Nicolas Mansard, CNRS and Ludo Righetti, NYU, Professor Abinav Valada, U. Bonn alongside a 20 strong Japanese delegation including JST officials. The workshop will provide an opportunity to exchange ideas and develop new connections besides leveraging existing mechanisms for International Project Collaborations defined within the Japanese Moonshot initiative. Further details to follow.

ICRA 2023 Workshop - Embracing contacts. Making robots physically interact with our world

A full day workshop bringing together world leading researchers from academia and industry focusing on contact-rich manipulation scenarios with expertise ranging from machine learning, mechanical design, control systems, planning and perception, to human robot close contact interaction will take place at ICRA2023. Workshop will take place 2 June 2023. [News item]

Edinburgh Science Festival 14April 2023

As part of the festival, we will open our labs to the general public to showcase a few of our robotic platforms and look at how robots can shape the future. The experience will open

from 9am until 5pm on Friday 14^th April 2023. This will be made up of one-hour slots with a maximum of 20 people in each group and booking will be open to adults or families with children over the age of 5. As the primary focus of our exhibition, we aim to create trust and confidence in robotic assistive systems rather than simply driving their technologies. By holding such exhibitions and allowing visitors to interact with robots, we hope to accelerate the process of using robots for assistive purposes in the future. [News item]

Meetings/presentations/engagement

Keynote at ICRA, 2023, 1 June 2023

Professor Sethu Vijayakumar delivered one of the ICRA 2023 Keynotes entitled 'From Automation to Autonomy: Machine Learning for Next-generation Robotics' on Thursday 1st June.

Distinguished Lectures on AI and Machine Learning, RIKEN-AIP

Professor Sethu Vijayakumar gave a Distinguished Lecture at RIKEN-AIP, Tokyo on 26 May 2023, titled “Robotics and AI for Assisted Living: Shared Autonomy in Action”

Invited panellist at AAAI/ACM Conference on AI, Ethics and Society

Dr Atoosa Kasirzadeh was keynote panellist on the panel: Large Language Models: Hype, Hope and Harm, August 2023.

Invited panellist at the 19^th Festival of Politics, hosted by the Scottish Parliament

Dr Atoosa Kasirzadeh was an invited panellist on “Where are the ethics in Artificial Intelligence” on the 11^th August 2023.

Invited speaker at Algorithmic Bias Workshop

Dr Atoosa Kasirzadeh was the invited speaker at this workshop at the University of Oxford in June 2023

Other activity

Dr Atoosa Kasirzadeh featured in Wall Street Journal story

Article titled “How Worried should we be about AI’s threat to Humanity? Even tech leaders can’t agree.” September 2023

How much of a threat is Artificial Intelligence? Inside Story

Dr Atoosa Kasirzadeh was a guest on this panel show discussing AI Safety.

Can regulating artificial intelligence suppress innovation? Inside Story

Dr Atoosa Kasirzadeh was a panel guest on this show discussing the AI act.

Publications

Daniel Gordon, Andreas Christou, Theodoros Stouraitis, Michael Gienger and Sethu Vijayakumar, Adaptive Assistive Robotics: A Framework for Triadic Collaboration Between Humans and Robots, Royal Society Open Science, vol. 10: 221617 (2023).
Juan Del Aguila Ferrandis, Joao Moura and Sethu Vijayakumar, Nonprehensile Planar Manipulation through Reinforcement Learning with Multimodal Categorical Exploration, Proc. IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS 2023), Detroit, USA (2023).
Ruaridh Mon-Williams, Theodoros Stouraitis and Sethu Vijayakumar, A Behavioural Transformer for Effective Collaboration between a Robot and a Non-stationary Human, Proc. IEEE International Conference on Robot and Human Interactive Communication (RO-MAN 2023), Busan, Korea (2023).
Christian Rauch, Ran Long, Vladimir Ivan and Sethu Vijayakumar, Sparse-Dense Motion Modelling and Tracking for Manipulation without Prior Object Models, IEEE Robotics and Automation Letters (RAL), vol. 7(4), pp. 11394-11401 (2022). (Presented at: IEEE International Conference on Robotics and Automation (ICRA 2023))
Daniel F. N. Gordon, Andreas Christou, Theodoros Stouraitis, Michael Gienger and Sethu Vijayakumar, Learning Personalised Human Sit-to-Stand Motion Strategies via Inverse Musculoskeletal Optimal Control, Proc. IEEE International Conference on Robotics and Automation (ICRA 2023), London, UK (2023).
Carlos Mastalli, Saroj Prasad Chhatoi, Thomas Corbères, Steve Tonneau and Sethu Vijayakumar,  Inverse-Dynamics MPC via Nullspace Resolution, IEEE Transactions on Robotics (T-RO), vol. 39(4), pp. 3222-3241 (2023). (Presented at: IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS 2023), Detroit, USA).
Henrique Ferrolho, Vladimir Ivan, Wolfgang Merkt, Ioannis Havoutis and Sethu Vijayakumar, RoLoMa: Robust Loco-Manipulation for Quadruped Robots with Arms, Autonomous Robots, vol. xx(yy), pp. zzz-zzz (2023).
Jaehyun Shim, Carlos Mastalli, Thomas Corbères, Steve Tonneau, Vladimir Ivan and Sethu Vijayakumar, Topology-Based MPC for Automatic Footstep Placement and Contact Surface Selection, Proc. IEEE International Conference on Robotics and Automation (ICRA 2023), London, UK (2023).
Abeba Birhane, Atoosa Kasirzadeh, David Leslie, Sandra Wachter, Science in the age of large language models, Nature Reviews Physics, April 2023
Atoosa Kasirzadeh, Iason Gabriel, In conversation with Artificial Intelligence: aligning language models with human values, Philosophy & Technology Vol 36, Springer June 2023 
Atoosa Kasirzadeh, Charles Evans, User Tampering in Reinforcement Learning Recommender Systems, Proc. AAAI/ACM Conference on AI, Ethics and Society, August 2023
Charlotte Bird, Eddie Ungless, Atoosa Kasirzadeh, Typology of Risks of Generative Text to Image Models, Proc. AAAI/ACM Conference on AI, Ethics and Society, August 2023
Katja Andrić, Atoosa Kasirzadeh, Reconciling Governmental Use of Online Targeting with Democracy, Proc. ACM Conference on Fairness, Accountability and Transparency, June 2023

Programme Updates October 2022 to March 2023

Highlights

Official launch of the Japan Moonshot Collaboration with Keynote and MoU signing between Turing and Waseda University (Moonshot Priority 3 Hub)
Five Turing affiliated papers from the Human-AI Interface and Robotics directorate accepted at ICRA 2023 in addition to high impact workshop on Close Contact Manipulation.
MoU signing with Digital University Kerala, India. DUK-UoE-Turing proposal submitted for State funding.

Strengths

Following on from the UK-Japan workshop, there has been a successful joint research funding with David Leslie and an ongoing collaboration with Tetsuya Ogata on one of the Japanese Governments Moonshot project. MoU signed 14 November 2022.
New collaboration through Moonshot project with Wellcome funded postdoc involving 6 months field study in Japan starting October 2022.

Opportunities

New Human-AI Interface Group Leader, Dr Atoosa Kasirzadeh, to lead on driving internal collaborations with Foundational Model researchers.

Upcoming plans/priorities

Develop showcase for Turing contribution to H2020 Harmony Project – Edinburgh Science Festival (Apr 2023), DUK symposium Kerala (Sep/Oct) and live demonstrations at Kawada, Japan (Nov 2023).
Follow up workshop for UK-Italy collaboration to be confirmed.

General programme updates

Upcoming events

Edinburgh Centre for Robotics/Alan Turing symposium 5^th June 2023

Title: Bridging International Robotics and AI Grand Challenges --Working with the Japanese Moonshot Initiative

ICRA 2023 Workshop - Embracing contacts. Making robots physically interact with our world

Edinburgh Science Festival 14April 2023

As part of the festival, we will open our labs to the general public to showcase a few of our robotic platforms and look at how robots can shape the future. The experience will open from 9am until 5pm on Friday 14^th April 2023. This will be made up of one-hour slots with a maximum of 30 people in each group and booking will be open to adults or families with children over the age of 5. As the primary focus of our exhibition, we aim to create trust and confidence in robotic assistive systems rather than simply driving their technologies. By holding such exhibitions and allowing visitors to interact with robots, we hope to accelerate the process of using robots for assistive purposes in the future. [News item]

Past Events

UK-Italy Robotics and AI research collaboration workshop

Follow up meeting took place on 23 March to discuss FAIR-ATI MoU and a joint bid for a UK-Italy UK Research activity in “trustworthy AI”.

Meetings/presentations/engagement

Robotics, AI and Pushing the Frontiers of Human Potential event

Professor Sethu Vijayakumar presented a talk ‘Robotics and the Future of AI’ at the event curated by TouchLab Limited in collaboration with the National Robotarium on Tuesday 31 January 2023. [News item]

Robotics and AI for Assisted Living: Shared Autonomy in Action

Future Robotics Organisation, Waseda University, invited Professor Sethu Vijayakumar to present a lecture on Monday 14^th November 2022 in Tokyo, Japan as part of the ongoing Moonshot programme. [News item]

Data Summit 2022, 3-4 November 2022

Professor Sethu Vijayakumar joined a panel at DataFest 2022 to discuss the topic ‘Human vs AI – who will win the skills battle?’ [News item]

The 8^th International Symposium on “Dialogue for Global Innovation” (DFGI-8), 5-6 October 2022

Recently held in London, the event focused on the theme “Technology and Humanity – Part 4: Humans need to apply ever-enhancing AI, including AI for Digital Health and Quantum Computing for the good of Society”. Professor Sethu Vijayakumar was an invited speaker at the event presenting his talk ‘Robotics and AI for Assisted Living: Shared Autonomy in Action’. [News item]

Other activity

The Alan Turing Institute-Waseda University MoU Signing

On the 14^th November 2022, Waseda University, Tokyo, signed a Memorandum of Understanding with the Alan Turing Institute, London in the presence of Professor Sethu Vijayakumar, Professor Tetsuya Ogata and Jun Ohya. This agreement will allow both organisations to officially start the Moonshot programme which will facilitate researcher and student exchange whilst promoting the establishment of future joint research projects. [News item]

University of Edinburgh-Alan Turing Institute-Digital University Kerala MoU Signing

On the 11^th October 2022 Kerala University signed a Memorandum of Understanding in the presence of Chief Minister Pinarayi Vijayan, Professor Sethu Vijayakumar, Professor Alex James and Professor Pankaj Pankaj. Following the State Government of Kerala’s announcement in 2022, DUK will be leading the establishment of a Digital Science Park and has invited the University of Edinburgh as the international academic partner to help set up this new initiative. [News item]

Dr Atoosa Kasirzadeh joined the Human-AI Interface as Group Leader. Atoosa’s research is focused on AI safety and ethics, including topics such as algorithmic transparency and accountability, algorithmic fairness and justice, ethical and social implications of generative AI models and recommender systems.

Wenqian Du started as a Postdoctoral Researcher affiliated to the Alan Turing Institute in January 2023. Two further Postdoctoral Researchers will start in April 2023.

Other research projects/activity

Ongoing Projects

Interactive collision-free bi-manual manipulation with dynamic obstacles without prior knowledge of the environment

Industry project funded by Kawada Robotics Corporation

PI – Professor Sethu Vijayakumar

Co-bots and Exoskeletons for Assisted Living with Ergonomic Measures

Industry project funded by Honda Research Institute Europe

PI – Professor Sethu Vijayakumar

Publications

Christopher Edwin Mower, Joao Moura, Nazanin Zamani Behabadi, Sethu Vijayakumar, Tom Vercauteren and Christos Bergeles, OpTas: An Optimization-based Task Specification Library for Trajectory Optimization and Model Predictive Control, IEEE International Conference on Robotics and Automation (ICRA 2023), London, UK (2023)

Namiko Saito, Joao Moura, Tetsuya Ogata, Marina Y. Aoyama, Shingo Murata, Shigeki Sugano and Sethu Vijayakumar, Structured Motion Generation with Predictive Learning: Proposing Subgoal for Long-Horizon Manipulation, IEEE International Conference on Robotics and Automation (ICRA 2023), London, UK (2023)

Christopher Mower, Theodoros Stouraitis, Joao Moura, Christian Rauch, Lei Yan, Nazanin Zamani Behabadi, Michael Gienger, Tom Vercauteren, Christos Bergeles and Sethu Vijayakumar, ROS Py-Bullet Interface: A Framework for Reliable Contact Simulation and Human-Robot Interaction, Conference on Robot Learning (CoRL 2022), Auckland, New Zealand (2022)

Andreas Christou, Daniel Gordon, Theodoros Stouraitis and Sethu Vijayakumar, Designing Personalised Rehabilitation Controllers using Offline Model-Based Optimisation, IEEE International Conference on Robotics and Biometrics (ROBIO 22), Xishuangbanna, China (2022) “IEEE Robio 2022 Best Paper Finalist”

Carlos Mastalli, Wolfgang Merkt, Josep Marti-Saumell, Henrique Ferrolho, Joan Sola, Nicolas Mansard and Sethu Vijayakumar, A Feasibility-Driven Approach to Control-Limited DDP, Autonomous Robots, vol. 46(8), pp. 985-1005 (2022)

Upcoming

New collaboration through the Moonshot program with Dr Giulia De Tongi (Wellcome funded postdoc in the University of Edinburgh). This research will involve a 6 month field study in Japan from October 2022 on the social acceptance of care robots.

Programme Updates April 2022 to September 2022

Highlights

Professor Sethu Vijayakumar visited India in July to widen the opportunities within the UK-India AI Partnership strategy announced in April. During this visit, Professor Vijayakumar visited Digital University Kerala to discuss the establishment of a Digital Science Park and Centre of Excellence.

Strengths

Following on from the UK-Japan workshop, there has been a successful joint research funding with David Leslie and an ongoing collaboration with Tetsuya Ogata on one of the Japanese Governments Moonshot project. Visiting research fellow hosted at the University of Edinburgh April – September 2022.
New collaboration through Moonshot project with Wellcome funded postdoc involving 6 months field study in Japan.

Opportunities

DUK-UoE-Turing collaboration on the establishment of a Digital science. MoU to be signed 11 October 2022 [News item]

Upcoming plans/priorities

DUK-UoE-Turing proposal to be submitted.

Follow up workshop for UK-Italy collaboration to be confirmed.

General programme updates

Upcoming events

ICRA 2023 Workshop - Embracing contacts. Making robots physically interact with our world

An application has been submitted to ICRA2023 for a full day workshop bringing together researchers from academia and industry focusing on contact-rich manipulation scenarios with expertise ranging from machine learning, mechanical design, control systems, planning and perception, to human robot close contact interaction. Turing will be the main sponsor of the workshop with a request for Out of Budget funding submitted. Decision expected 21 December 2022. Workshop will take place 29 May 2023 or 2 June 2023. [News item]

UK-Italy Robotics and AI research collaboration workshop

There will be a follow up workshop held in London. Details to follow.

Edinburgh Science Festival 14April 2023

As part of the festival we will open our labs to the general public to showcase a few of our robotic platforms and look at how robots can shape the future. The experience will open from 9am until 5pm on Friday 14^th April 2023. This will be made up of one hour slots with a maximum of 20 people in each group and booking will be open to adults or families with children over the age of 5. As the primary focus of our exhibition, we aim to create trust and confidence in robotic assistive systems rather than simply driving their technologies. By holding such exhibitions and allowing visitors to interact with robots, we hope to accelerate the process of using robots for assistive purposes in the future. [News item]

Meetings/presentations/engagement

The Turing Lectures: Where next for self-driving vehicles? 27 September 2022

In this Turing lecture hosted by Professor Sethu Vijayakumar, Professor Sarah Sharples (Chief Scientific Adviser for the Department for Transport) discussed the interface between policy and science for self-driving vehicles and highlighted some of the challenges and opportunities that we face in this area. [News item]

AgriForwards CDT conference, Lincoln, 22 June 2022

Associate Professor Nick Hawes was invited to present a talk at the University of Lincoln.

KTH RPL Summer School, Sweden, 14 June 2022

Associate Professor Nick Hawes was invited to present a talk titled ‘Better Autonomy through Uncertainty’ at the recent KTH RPL Summer School.

Dev<Talks/>, 8 June 2022

Professor Sethu Vijayakumar was invited to present the opening keynote of Dev<Talks/> in Bucharet, Romania. The talk was entitled ‘From Automation to Autonomy: AI driving the Future of Interactive Robotics. [News item]

AWS sponsored conference, Oxford, 28 May 2022

Associate Professor Nick Hawes was invited to present a talk on human-machine collaboration.

Altair executive dinner, 23 May 2022

Professor Sethu Vijayakumar invited keynote speaker and panellist at Altair Executive Dinner and Networking event. Discussion took place on the topic ‘Exploiting the Age of Digital’. [News item]

Vision Zero Summit Japan 2022, 11-13 May 2022

Consisting of 16 technical sessions presenting innovative solutions to a wide range of topics including redefining safety, health and wellbeing for the New Normal, Professor Sethu Vijayakumar was invited to present his findings on ‘Shared Autonomy for Dependable Human Robot Interactions’ as part of the session ‘AI/ICT & digitization’. [News item]

Adapt research centre Dublin, 28 April 2022

Professor Sethu Vijayakumar was invited to present a talk at the world-leading SFI Research Centre for AI-Driven Digital Content Technology in Ireland. The talk was entitled ‘From Automation to Autonomy: Novel Machine Learning driving the Next-Generation of Robotics’. [News item]

Edinburgh Centre for Robotics, 12 April 2022

Associate Professor Nick Hawes presented a talk entitled ‘Mission Planning with Uncertain Models’ at the recent Gateway Event

Other activity

Three new Postdoctoral Researchers are due to start early 2023 affiliated to the Alan Turing Institute.

Dr Namiko Saito, Visiting Research Fellow from Waseda University visited the University of Edinburgh from April – September 2022 as part of the Japanese Moonshot project. Dr Saito spent her time working with the Nextage robot on generalizable sequencing of close contact interaction skills with a dual-arm robot while co-supervising and collaborating with existing UoE funded and Turing affiliated PhD student, Marina Aoyama, on a project looking at task/object transfer of skills in real time.

Other research projects/activity

Ongoing Projects

Interactive collision-free bi-manual manipulation with dynamic obstacles without prior knowledge of the environment

Industry project funded by Kawada Robotics Corporation

PI – Professor Sethu Vijayakumar

Co-bots and Exoskeletons for Assisted Living with Ergonomic Measures

Industry project funded by Honda Research Institute Europe

PI – Professor Sethu Vijayakumar

Publications

Jiayi Wang, Teguh Santoso Lembono, Sanghyun Kim, Sylvain Calinon, Sethu Vijayakumar and Steve Tonneau, Learning to Guide Online Multi-Contact Receding Horizon Planning, IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS 2022), Kyoto, Japan (2022).

Traiko Dinev, Carlos Mastalli, Vladimir Ivan, Steve Tonneau and Sethu Vijayakumar, A Versatile Co-Design Approach For Dynamic Legged Robots, IEEE/RSJ Intl. Conf. On Intelligent Robots and Systems (IROS 2022), Kyoto, Japan (2022).

Ran Long, Christian Rauch, Tianwei Zhang, Vladimir Ivan, Tin Lun Lam and Sethu Vijayakumar, RGB-D SLAM in Indoor Planar Environments with Multiple Large Dynamic Objects, IEEE Robotics and Automation Letters (RAL, vol. 7(3), pp. 8209-8216 (2022).

Upcoming

New collaboration through the Moonshot program with Dr Giulia De Tongi (Wellcome funded postdoc in the University of Edinburgh). This research will involve a 6 month field study in Japan from October 2022 on the social acceptance of care robots.
Professor Sethu Vijayakumar due to visit Japan in November 2022 where he will present a talk at Waseda University as part of the ongoing Moonshot Programme and attend an official MoU signing.

Programme Updates October 2021 to March 2022

Highlights

UK-Italy Robotics and AI research collaboration workshop recently held in Rome, Italy.

UK-India: Towards a Digital Future workshop held at Wilton Park.

Strengths

Following on from the UK-Japan workshop, there has been a successful joint research funding with David Leslie and an ongoing collaboration with Tetsuya Ogata on one of the Japanese Governments Moonshot project. Visiting research fellow to be hosted at the University of Edinburgh April – September 2022.

New collaboration through Moonshot project with Wellcome funded postdoc involving 6 months field study in Japan.

Opportunities

UK-India AI Partnership

Challenges and risks

Ongoing issues with the international spread of the coronavirus (COVID-19).

Upcoming plans/priorities

Follow up workshop for UK-Italy collaboration to take place in autumn 2022.

UK-India AI Partnership opportunities – Professor Sethu Vijayakumar to visit India in July.

General programme updates

Past events

IEEE UK and Ireland RAS Chapter Conference 2022 23 February 2022

The annual conference took place in Edinburgh with over 80 participants attending. This interactive event was targeted at robotics practitioners, PhD students, PDRAs, researchers in academia and industry professionals to discuss blue sky, high-risk moonshot ideas. The event included a poster session, short selective talks by invited speakers in the four key themed areas of RAS and an interactive panel interspersed with inspiring keynotes by leading UK researchers. [News item]

UK-Italy Robotics and AI research collaboration workshop 1^st March 2022

The workshop took place at the residence of the British Ambassador, Villa Wolkonsky in Rome bringing together leading researchers in AI and Robotics from Italy and the UK. The Italian organisations involved in this activity are CINI-AIIS (Consorzio Interuniversitario Nazionale per Informatica – Laboratorio Nazionale di Artificial Intelligence and Intelligent Systems), Artes 4.0 (Centro di Competenza ad Alta Specializzazione in Industria 4.0 e Robotica) and IIT (Istituto Italiano di Tecnologia). The first workshop was a basis for identifying the existing collaboration landscape and research priority topics for future joint funding opportunities. [News item]

India and the UK: Towards a Digital Future workshop 9-11 March 2022

The workshop took place at Wilton Park, the executive agency of the UK Foreign, Commonwealth & Development Office (FCDO). Over 40 stakeholders from business, academia and government from both the UK and India attended the event to refine the ‘why’ and ‘what’ of the UK-India AI partnership proposal. 

Upcoming events

UK-Italy Robotics and AI research collaboration workshop Autumn 2022

There will be a follow up workshop held in London later in 2022. Details to follow.

Meetings/presentations/engagement

Releasing Robots into the Wild Virtual Workshop 15^th October 2021

Professor Nick Hawes presented a talk on ‘Experiences of Deploying Autonomous Service Robots in Everyday Environments’ as part of the virtual workshop ‘Releasing Robots into the Wild’. YouTube Link

Atlas Talk University of Manchester 12 January 2022

Professor Nick Hawes presented a talk to at the University of Manchester entitled ‘Mission planning with uncertain models’. In the talk Professor Hawes presented a high-level summary of his recent work ranging from planning in Markov decision processes (MDPs) with a Gaussian process over a single state feature, to planning in uncertain MDPs and Bayes-adaptive MDPs where the true model cannot be known with certainty.

India Science Festival 21 January 2022

Professor Sethu Vijayakumar presented a talk entitled ‘Future of Automated Robots: How far have we come?’ at the recent India Science Festival where attendees had the opportunity to hear about the evolution of automated robots over the years and were given the opportunity to watch a live demo of a humanoid robot, quadruped robot and the Nextage dual-arm robot streamed directly from the Edinburgh laboratory. [News item]

CompSoc Student Society 18 February 2022

Professor Sethu Vijayakumar was invited to present a talk on ‘From Automation to Autonomy: Data-driven Machine Learning for Interactive Robots’ to an audience of undergraduate and postgraduate computer science students. [News item]

Beijing Institute of Technology 24^th February 2022

Professor Nick Hawes was invited to present an online talk at the Beijing Institute of Technology.

Science AAAS Webinar 9^th March 2022

‘Personal AI-based robots as lifetime human companions’. Science AAAS welcomed Professor Sethu Vijayakumar, alongside other experts at Waseda University, to share the latest technology developments that give us some insight into what a future world might look like. [News item]

Other activity

The Future of Robotics WIRED documentary

Professor Sethu Vijayakumar provided his insight into the future of robotics in a four part WIRED documentary.

Part 1 – How much control should we give robots? Link

Part 2 – Shared Autonomy & Work Link

Part 3 – How Exoskeleton Technology Will Transform Healthcare Link

Part 4 – What Robots & AI Do With Your Data Link

Professor Nick Hawes led a Spot inspection Demo on the 26^th January 2022 to an international delegation from Chernobyl, 16^th March 2022 at UKAEA/RACE for RAIN final showcase and also on the 30^th March 2022 at ISCF Robots for a Safer World event.

Dr Saeid Samadi appointed new Post-Doctoral Researcher Associate. Start date 21^st March 2022.

Professor Nick Hawes invited to present a seminar to AIMS CDT on 12^th November 2021.

Other research projects/activity

Ongoing Projects

Interactive collision-free bi-manual manipulation with dynamic obstacles without prior knowledge of the environment

Industry project funded by Kawada Robotics Corporation

PI – Professor Sethu Vijayakumar

Co-bots and Exoskeletons for Assisted Living with Ergonomic Measures

Industry project funded by Honda Research Institute Europe

PI – Professor Sethu Vijayakumar

Publications

Daniel F.N. Gordon, Christopher McGreavy, Andreas Christou and Sethu Vijayakumar, Human-in-the-loop Optimization of Exoskeleton Assistance via Online Simulation of Metabolic Cost, IEEE Transactions on Robotics (T-RO), IEEE International Conference on Robotics and Automation (ICRA 2022)

Shen Li, Theodoros Stouraitis, Michael Gienger, Sethu Vijayakumar and Julie Shah, Set-based State Estimation with Probabilistic Consistency Guarantee under Epistemic Uncertainty, IEEE Robotics and Automation Letters (RAL), vol. 7(3), pp. 5958-5965 (2022).

Generic Papers publication list

Christian Rauch, Ran Long, Vladimir Ivan and Sethu Vijayakumar, Sparse-Dense Motion Modelling and Tracking for Manipulation without Prior Object Models, IEEE Robotics and Automation Letters (RAL) 2022.

Joao Moura, Theodoros Stouraitis and Sethu Vijayakumar, Non-prehensile Planar Manipulation via Trajectory Optimization with Complementarity Constraints, Proc. IEEE International Conference on Robotics and Automation (ICRA 2022), Philadelphia, USA (2022).

Upcoming

Dr Namiko Saito, Visiting Research Fellow from Waseda University will be based at the University of Edinburgh from April – September 2022 as part of the Japanese Moonshot project. Dr Saito will be working with the Nextage robot on generalizable sequencing of close contact interaction skills with a dual-arm robot.
Dr Namiko Saito will co-supervise and collaborate with existing UoE funded and Turing affiliated PhD student, Marina Aoyama, on a project looking at task/object transfer of skills in real time.
New collaboration through the Moonshot program with Dr Giulia De Tongi (Wellcome funded postdoc in the University of Edinburgh). This research will involve a 6 month field study in Japan from October 2022 on the social acceptance of care robots.
Professor Sethu Vijayakumar due to visit India in July to widen the opportunities within the UK-India AI Partnership strategy which is to be announced in April. During this visit, Professor Vijayakumar will visit IIT (Palakkad, Chennai, and Bangalore) to interact with their AI institutes and also explore a UKIERI grant writing opportunity with National Institute of Technology, Trichy.
Professor Sethu Vijayakumar due to visit Japan in November 2022 where he will present a talk at Waseda University as part of the ongoing Moonshot Programme.

Finished/funded completed projects

Project title – Model Criticism in Multi-Agent Systems Project, Ended October 2021

PI – Dr Stefano Albrecht

Programme Updates April 2021 to September 2021

Highlights

Collaboration with Tetsuya Ogata on one of the Japanese Governments Moonshot Projects ‘Smart Robot that is Close to One Person for a Lifetime’.

Strengths

Involvement in the Japan AI Moonshot projects.

Opportunities

UK-Italy Robotics and AI Collaboration Workshop to be held in February 2022.

India and the UK collaboration event at Wilton Park (date TBC).

Challenges and risks

Number of events postponed or cancelled due to the international spread of the coronavirus (COVID-19).

Upcoming plans/priorities

Seek support from BEIS, OAI, UKRI and on Japanese side (METI, MEXT) on broader project proposal outcomes from the UK-Japan workshop held in September 2019.

General programme updates

Past Events

Alan Turing Multi-Agent Systems Seminar Series 22 April 2021

Professor Nick Hawes joined Dr Bruno Lacerda to present the talk ‘Towards Guaranteed Multi-Robot Systems under Uncertainty’. In the talk they presented their vision for how to plan the behaviour of multi-robot systems under uncertainty and discuss approaches for generating system behaviour which delivers probabilistic guarantees. Drawing the distinction between guarantees on team behaviour and individual robot behaviour.

Atom Society Talk 20 May 2021

Professor Nick Hawes presented a talk to the Atom Society entitled ‘An Uncertain Mission: Decision Making for Robots’. In the talk Professor Hawes presented his recent work on mission planning algorithms that model the uncertainty of robot actions and spoke about how algorithms have been applied to control robots in a range of domains from greeting people in a hospital to inspecting nuclear waste stores.

Automatica Sprint 2021 22 June 2021

The prestigious Automatica Hightech Summit recently took place on the 22 June 2021 where leading international masterminds from the field of AI and robotics were invited to discuss their visions, innovations, insights and theories. Invited as a keynote speaker as part of the session ‘Beyond Human Spaces’, Professor Sethu Vijayakumar presented his findings on “Shared Autonomy: The Future of Interactive Robotics”. [News item]

UKRI TAS Resilience Node

On the 5^th July 2021, Professor Nick Hawes presented an online talk on Resilience: Mission Planning for Long-Term Autonomy in Mobile Service Robots.

AI Research in Japan: Challenges at the RIKEN Center for Advanced Intelligence Project

On the 13^th July 2021, Professor Sugiyama Masashi joined Professor Sethu Vijayakumar in a live event to introduce the latest research conducted at the RIKEN AIP and its recent achievements and challenges. Participants were given the opportunity to learn more about Japan’s latest research into AI technology and its real-world applications with a live Q&A session held during the event.

This event was co-organised by The Alan Turing Institute and The Embassy of Japan in UK and hosted by Japan House London. [News item]

RSS 2021 Workshop

Professor Nick Hawes presented a talk on Declarative and Neurosymbolic Representations in Robot Learning and Control: Mission Planning with Uncertain Models at the recent RSS 2021 workshop held on 15^th July 2021.

TAROS 2021 IET Lecture – An Uncertain Mission: Decision Making for Autonomous Robots

On the 9^th September 2021, Professor Nick Hawes delivered the online IET TAROS public lecture.

Upcoming events

IEEE UK RAS UK & I Conference Wednesday 23^rd February 2022

The annual IEEE UK & I RAS conference will be held at the University of Edinburgh hosted by Professor Sethu Vijayakumar.  The aim of the conference is to improve the communication between members, researchers, students and industrial professionals who are interested in the activities of RAS in research, development and education. This event is aimed at the next generation of roboticist and combines a poster session with a series of talks by leading UK researchers. [News item]

UK-Italy Robotics and AI Collaboration Workshop February 2022

Led by Professor Angelo Cangelosi. Aim to explore collaboration opportunities between the two countries. The Italian organisations initially involved in this activity are CINI-AIIS (Consorzio Interuniversitario Nazionale per Informatica - Laboratorio Nazionale di Artificial Intelligence and Intelligent Systems), Artes 4.0 (Centro di Competenza ad Alta Specializzazione in Industria 4.0 e Robotica) and IIT (Istituto Italiano di Tecnologia). The workshop will bring together leading researchers in AI and robotics from Italy and the UK. Further Information

India and the UK: Towards a Digital Future Expected late 2021/ early 2022

Wilton Park is an executive agency of the UK Foreign, Commonwealth & Development Office (FCDO), supporting British foreign policy priorities which are core to the UK’s public diplomacy work. Wilton Park will be partnering with the BHC in New Delhi to hold an event on AI with a strong focus on healthcare co-badged with Alan Turing.

Other activity

Professor Nick Hawes joined a team presenting to the public at the Goodwood Festival of Speed.

Dr Chris Mower appointed new Post-Doctoral Research Associate. Start date 20 September 2021.

Professor Nick Hawes co-organiser of IJCAI R2AW workshop

Professor Nick Hawes led Oxford’s UK-RAS online event for general public (approx. 100 attendees).

Publications

Jiayi Wang, Sanghyun Kim, Sethu Vijayakumar and Steve Tonneau, Multi-Fidelity Receding Horizon Planning for Multi-Contact Locomotion, Proc. IEEE-RAS 20^th International Conference on Humanoid Robots (Humanoids 2020), Munich, Germany (2021)

Chris Mower, Joao Moura and Sethu Vijayakumar, Skill-based Shared Control, Proc. Robotics: Science and Systems (R:SS 2021)

Henrique Ferrolho, Wolfgang Merkt, Carlo Tiseo and Sethu Vijayakumar, Residual Force Polytope: Admissible Task-Space Forces of Dynamic Trajectories, Robotics and Autonomous Systems (2021)

Tianwei Zhang, Huayan Zhang, Xiaofei Li, Junfeng Chen, Tin Lum Lam and Sethu Vijayakumar, AcousticFusion: Fusing Sound Source Localization to Visual SLAM in dynamic environments, Proc. IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS 2021)

Huayan Zhang, Tianwei Zhang, Tin Lun Lam and Sethu Vijayakumar, PoseFusion2: Simultaneous Background Reconstruction and Human Shape Recovery in Real-time, Proc. IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS 2021)

Carlo Tiseo, Wolfgang Merkt, Keyhan Kouhkiloui Babarahmati, Wouter Wolfslag, Ionnis Havoutis, Sethu Vijayakumar and Michael Mistry, HapFIC: An Adaptive Force/Position Controller for Safe Environment Interaction in Articulated Systems, IEEE Transactions on Neural Systems and Rehabilitation Engineering (IEEE-TNSRE), vol. 29, pp. 1432-1440 (2021)

Wolfgang Merkt, Vladimir Ivan, Traiko Dinev, Ioannis Havoutis and Sethu Vijayakumar, Memory Clustering using Persistent Homology for Multimodality and Discontinuity-Sensitive Learning of Optimal Control Warm Starts, IEEE Transactions on Robotics (T-RO) (IROS 2021)

Model criticism in Multi-Agent Systems Project

Project Duration: 01/01/2020 – 31/12/2021

PI – Dr Stefano V. Albrecht

Project Website

Status of progress against goals, any blockers to progress

All project goals set out in the original proposal have been achieved. The outcomes associated to the two main objectives in proposal were published in the papers below.

Josiah P. Hanna, Arrasy Rahman, Elliot Fosong, Francisco Eiras, Mihai Dobre, John Redford, Subramanian Ramamoorthy, Stefano V. Albrecht Interpretable Goal Recognition in the Presence of Occluded Factors for Autonomous Vehicles IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021 https://arxiv.org/abs/2108.02530

Ibrahim H. Ahmed, Josiah P. Hanna, Elliot Fosong, Stefano V. Albrecht Towards Quantum-Secure Authentication and Key Agreement via Abstract Multi-Agent Interaction International Conference on Practical Applications of Agents and Multi-Agent Systems (PAAMS), 2021 https://arxiv.org/abs/2007.09327

Papers submitted or published

[1] Stefano V. Albrecht, Cillian Brewitt, John Wilhelm, Balint Gyevnar, Francisco Eiras, Mihai Dobre, Subramanian Ramamoorthy Interpretable Goal-based Prediction and Planning for Autonomous Driving IEEE International Conference on Robotics and Automation (ICRA), 2021 https://arxiv.org/abs/2002.02277

[2] Cillian Brewitt, Balint Gyevnar, Samuel Garcin, Stefano V. Albrecht GRIT: Fast, Interpretable, and Verifiable Goal Recognition with Learned Decision Trees for Autonomous Driving IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021 https://arxiv.org/abs/2103.06113

[3] Josiah P. Hanna, Arrasy Rahman, Elliot Fosong, Francisco Eiras, Mihai Dobre, John Redford, Subramanian Ramamoorthy, Stefano V. Albrecht Interpretable Goal Recognition in the Presence of Occluded Factors for Autonomous Vehicles IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021 https://arxiv.org/abs/2108.02530

[4] Henry Pulver, Francisco Eiras, Ludovico Carozza, Majd Hawasly, Stefano V. Albrecht, Subramanian Ramamoorthy PILOT: Efficient Planning by Imitation Learning and Optimisation for Safe Autonomous Driving IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021 https://arxiv.org/abs/2011.00509

[5] Francisco Eiras, Majd Hawasly, Stefano V. Albrecht, Subramanian Ramamoorthy A Two-Stage Optimization-based Motion Planner for Safe Urban Driving IEEE Transactions on Robotics (T-RO), 2021 https://arxiv.org/abs/2002.02215

[6] Ibrahim H. Ahmed, Josiah P. Hanna, Elliot Fosong, Stefano V. Albrecht Towards Quantum-Secure Authentication and Key Agreement via Abstract Multi-Agent Interaction International Conference on Practical Applications of Agents and Multi-Agent Systems (PAAMS), 2021 https://arxiv.org/abs/2007.09327

Leveraged funding/support

£50k from the UoE "Integrative AI" Turing funds led by Michael Rovatsos. Project is titled "Asking Your Autonomous Car to Explain its Decisions: Towards AI That Explains Itself" and supports one full-time postdoc for 1 year.

Currently leveraging additional funds from Mirella Lapata's ELIAI grant to extend this project, and in conversation with Huawei for additional funding.

Programme Updates January 2021 to March 2021

Highlights

Collaboration with Tetsuya Ogata on one of the Japanese Governments Moonshot Projects ‘Smart Robot that is Close to One Person for a Lifetime’.

Strengths

Involvement in the Japan AI Moonshot projects.

Opportunities

UK-Italy Robotics and AI Collaboration Workshop to be held in 2021.

Challenges and risks

Number of events postponed or cancelled due to the international spread of the coronavirus (COVID-19).

Upcoming plans/priorities

Seek support from BEIS, OAI, UKRI and on Japanese side (METI, MEXT) on broader project proposal outcomes from the UK-Japan workshop held in September 2019.

General programme updates

Past Events

Scotsman Inside Science: Robots can lend a personal touch to human healthcare

In a recent article published in The Scotsman, Professor Sethu Vijayakumar looks at the potential of robotics in healthcare and explains how the latest advancements could be used in the fight against cancers or lung damage brought on by Covid-19. [News item]

Leading Britain’s Conversation LBC

Professor Sethu Vijayakumar was recently interviewed by Ruth Davidson in a podcast ‘An Inconvenient Ruth’.

Autonomy and Verification Network

Professor Nick Hawes presented at the Workshop on Verification and Regulatory Issues for Remote Robotic Inspection.

Upcoming events

Munich-I High-tech Summit, Automatica Sprint Tuesday 22^nd June 2021

Professor Sethu Vijayakumar has been invited as a keynote speaker at the Automatica Sprint in Munich as part of the session ‘Beyond Human Spaces. Professor Vijayakumar will be speaking on the subject ‘Shared Autonomy: The Future of Interactive Robots’.

IEEE UK RAS UK & I Conference Wednesday 27^th January 2021 *Postponed – date TBC*

Edinburgh International Science Festival April 2021 *TBC*

There will be a Talos and Nextage demonstration carried out during the2021 International Festival.

UK-Italy Robotics and AI Collaboration Workshop September 2021

Other activity

Jaehyun Shim has been appointed as Laboratory Software Engineer co-badged with Alan Turing. Jaehyun will join the Edinburgh team from June 2021.
Post Doctoral Research Associate position advertised co-badged with Alan Turing. Closing date 5^th May 2021.
Professor Nick Hawes appointed as MPLS Academic Champion for Public Engagement with Research.

Publications

Ran Long, Christian Rauch, Tianwei Zhang, Vladimir Ivan and Sethu Vijayakumar, RigidFusion: Robot Localisation and Mapping in Environments with Large Dynamic Rigid Objects, IEEE Robotics and Automation Letters (RAL), 2021

Lei Yan, Theodorous Stouraitis and Sethu Vijayakumar, Decentralized Ability-Aware Adaptive Control for Multi-robot Collaborative Manipulation, IEEE Robotics and Automation Letters (RAL), 2021

Guiyang Xin, Songyan Xin, Oguzhan Cebe, Mathew Jose Pollayil, Franco Angelini, Manolo Garabini, Sethu Vijayakumar and Michael Mistry, Robust Footstep Planning and LQR Control for Dynamic Quadrupedal Locomotion, IEEE Robotics and Automation Letters (RAL), 2021 

Henrique Ferrolho, Vladimir Ivan, Wolfgang Xaver Merkt, Ioannis Havoutis and Sethu Vijayakumar, Inverse Dynamics vs. Forward Dynamics in Direct Transcription Formulations for Trajectory Optimization, Proc. IEEE International Conference on Robotics and Automation (ICRA 2021), Xian, China, 2021

Xiaoyu Zhang, Lei Yan, Tin Lun Lam and Sethu Vijayakumar, Task-Space Decomposed Motion Planning Framework for Multi-Robot Loco-Manipulation, Proc. IEEE International Conference on Robotics and Automation (ICRA 2021), Xian, China, 2021

Traiko Dinev, Wolfgang Xaver Merkt, Vladimir Ivan, Ioannis Havoutis and Sethu Vijayakumar, Sparsity-Inducing Optimal Control via Differential Dynamic Programming, Proc. IEEE International Conference on Robotics and Automation (ICRA 2021), Xian, China, 2021

Jiatao Ding, Songyan Xin, Tin Lun Lam and Sethu Vijayakumar, Versatile Locomotion by Integrating Ankle, Hip, Stepping, and Height Variation Strategies,  Proc. IEEE International Conference on Robotics and Automation (ICRA 2021), Xian, China, 2021

Carlo Tiseo, Vladimir Ivan, Wolfgang Xaver Merkt, Ioannis Havoutis, Michael Mistry and Sethu Vijayakumar, A Passive Navigation Planning Algorithm for Collision-free Control of Mobile Robots, Proc. IEEE International Conference on Robotics and Automation (ICRA 2021), Xian, China, 2021

Ewen Dantec, Rohan Budhiraja, Adria Roig, Teguh Santoso Lembono, Guilhem Saurel, Olivier Stasse, Pierre Fernbach, Steve Tonneau, Sylvain Calinon, Sethu Vijayakumar, Michel Taix and Nicolas Mansard, Whole Body Model Predictive Control with Memory of Motion: Experiments on a Torque-Controlled TALOS, Proc. IEEE International Conference on Robotics and Automation (ICRA 2021), Xian, China (2021)

Model criticism in Multi-Agent Systems Project

Project Duration: 01/01/2020 – 31/12/2021

PI – Dr Stefano V. Albrecht

Project Website

Status of progress against goals, any blockers to progress

Started new collaboration between the lab and Dematic on a project about applying multi-agent reinforcement learning to multi-robot warehouse management. Dematic are a large global player in this area, owned by KION. First phase of the project (4 months) has ended now with promising results. They want to expand the project into next phase and discussions on funding and resources are underway.

Status of progress against goals, any blockers to progress

Progress is good, no blockers

Papers submitted or published

Paper published at ICRA 2021: Stefano V. Albrecht, Cillian Brewitt, John Wilhelm, Balint Gyevnar, Francisco Eiras, Mihai Dobre, Subramanian Ramamoorthy Interpretable Goal-based Prediction and Planning for Autonomous Driving IEEE International Conference on Robotics and Automation (ICRA), 2021 https://arxiv.org/abs/2002.02277

Papers submitted to IROS 2021: Cillian Brewitt, Balint Gyevnar, Stefano V. Albrecht GRIT: Verifiable Goal Recognition for Autonomous Driving using Decision Trees arXiv, 2103.06113, 2021 https://arxiv.org/abs/2103.06113

Josiah Hanna, Elliot Fosong, Francisco Eiras, Mihai Dobre, Subramanian Ramamoorthy, Stefano Albrecht Interpretable Goal Recognition in the Presence of Hidden Obstacles for Autonomous Vehicles  Henry Pulver, Francisco Eiras, Ludovico Carozza, Majd Hawasly, Stefano Albrecht, Subramanian Ramamoorthy PILOT: Efficient Planning by Imitation Learning and Optimisation for Safe Autonomous Driving 

Papers published in ALA workshop at AAMAS 2021: Arrasy Rahman, Niklas Höpner, Filippos Christianos, Stefano V. Albrecht Towards Open Ad Hoc Teamwork Using Graph-based Policy Learning AAMAS Workshop on Adaptive and Learning Agents, 2021 Filippos Christianos, Georgios Papoudakis, Arrasy Rahman, Stefano V. Albrecht Scaling Multi-Agent Reinforcement Learning with Selective Parameter Sharing AAMAS Workshop on Adaptive and Learning Agents, 2021 Georgios Papoudakis, Filippos Christianos , Lukas Schäfer, Stefano V. Albrecht Comparative Evaluation of Multi-Agent Deep Reinforcement Learning Algorithms AAMAS Workshop on Adaptive and Learning Agents, 2021

Articles/blogs/media/press releases (include links)

Project blog

External engagement, influence, impact (Academic/industry/government/public/international)

Started monthly Multi-Agent Systems Talk Series as part of the MAS special interest group. First talk by Allan Dafoe (Oxford) and Edward Hughes (DeepMind), second talk by Kobi Gal (Edinburgh). More speakers for coming months already confirmed. Very good attendance, 70-90 people. Speaker list will be published soon on group page (https://www.turing.ac.uk/research/interest-groups/multi-agent-systems) and recordings will be uploaded to a YouTube channel (in progress).

Leveraged funding/support

Planning to apply to two grants: - TAS Hub prime funding - US DoD grant together with Roke Manor (UK contractor)

Programme Updates - October 2020 to December 2020

Opportunities

UK-Italy Robotics and AI Collaboration Workshop to be held in 2021.

Involvement with the Japan AI moonshot projects.

Challenges and risks

Number of events postponed or cancelled due to the international spread of the coronavirus (COVID-19).

Upcoming plans/priorities

Seek support from BEIS, OAI, UKRI and on Japanese side (METI, MEXT) on broader project proposal outcomes from the UK-Japan workshop held in September 2019.

General programme updates

Past Events

Cambridge Consultants Roundtable Tokyo 14^th October 2020

Professor Sethu Vijayakumar presented in the Collaborations in AI roundtable workshop organised by the British Embassy in Tokyo.

Professor Nick Hawes invited to talk at ICAPS 20 Summer School - Mission Planning with Learned Models for Long-Term Autonomy.

Professor Nick Hawes invited to talk at IROS 20 Long-Term Autonomy Workshop - Mission Planning with Learned Models for Long-Term Autonomy.

Upcoming events

IEEE UK RAS UK & I Conference Wednesday 27^th January 2021 *Postponed – date TBC*

Edinburgh International Science Festival April 2021

There will be a Talos and Nextage demonstration carried out during the2021 International Festival.

UK-Italy Robotics and AI Collaboration Workshop September 2021

Other activity

One Postdoctoral Researcher and one Laboratory Software Engineer position co-badged with Alan Turing advertised.

Professor Nick Hawes presented collaborative work with Accenture Labs as part of the Accenture/ATI Innovation Symposium 2020.

Model criticism in Multi-Agent systems Project

Project Duration: 01/01/2020 – 31/12/2021

PI – Dr Stefano V. Albrecht

Project website

Status of progress against goals, any blockers to progress

For thrust in autonomous driving:

two papers currently under review at ICRA
one paper in preparation for IROS

For thrust in secure authentication:

Going through some conference cycles (NeurIPS, IAAI, both not successful). Now consolidating feedback and preparing for submission to an upcoming security conference (tbd).

External engagement, influence, impact (Academic/industry/government/public/international)

Turing MAS group - setting up monthly talk schedule.

Papers submitted or published

F. Christianos, L. Schäfer, S.V. Albrecht Shared Experience Actor-Critic for Multi-Agent Reinforcement Learning Conference on Neural Information Processing Systems (NeurIPS), 2020 https://arxiv.org/abs/2006.07169

Software/code/tools/methods developed/released

Released new software at public code repository: https://github.com/uoe-agents Includes implementation of algorithm and environments proposed in NeurIPS 2020 paper.

Articles/blogs/media/press releases (include links)

We have a blog here.

Leveraged funding/support

Other research projects/activity

Further Postdoctoral Researcher positions to be advertised co-badged with Alan Turing.

Professor Nick Hawes part of the team awarded a new EPSRC programme grant.

Programme Updates - April 2020 to September 2020

Highlights

Building Dynamic Robots lecture with Boston Dynamics

Opportunities

UK-Italy Robotics and AI Collaboration Workshop to be held in 2021.

Involvement with the Japan AI moonshot projects.

Challenges and risks

Number of events postponed or cancelled due to the international spread of the coronavirus (COVID-19).

Upcoming plans/priorities

Seek support from BEIS, OAI, UKRI and on Japanese side (METI, MEXT) on broader project proposal outcomes from the UK-Japan workshop held in September 2019.

General programme updates

Past Events

Professor Sethu Vijayakumar and Marc Railbert

Building Dynamic Robots with Boston Dynamics Tuesday 22^nd September 2020

With over 630 delegates watching online, Professor Sethu Vijayakumar led a conversation with Marc Raibert, the man behind one of the largest robotics companies in the world. Discussion took place on the latest robots currently being worked on at Boston Dynamics, new robots being developed and a live demonstration of a robot being operated across continents (Edinburgh to Boston). A Q&A session followed. [News item]

RAIN Webinar series Thursday 24^th September 2020

Professor Nick Hawes presented an overview of the work in the UKRI hub on Remote Inspection

Jericho Roundtable discussion Friday 4^th September 2020

Professor Sethu Vijayakumar joined the Jericho roundtable to discuss the role of robotics as an enabler to support the upscaling of the workforce and increase economic productivity over the long term and the role of robots across the healthcare, manufacturing and service sectors.

World Artificial Intelligence Conference Friday 10^th July 2020

Invite World Artificial Intelligence Conference — World Artificial Intelligence Conference

Professor Sethu Vijayakumar recently spoke at the Robot and Artificial Intelligence industry forum with the theme ‘Cloud AI Empowering Robots – Digital Finance Boosting Revolution’. [News item]

RSE Artificial Intelligence event Tuesday 30^th June 2020

The Royal Society of Edinburgh hosted a series of short online talks and discussion on the topic of Artificial Intelligence. Professor Sethu Vijayakumar presented a live short talk on the topic ‘From Automation to Autonomy: Collaborative Robots driving the Future of Work’. [News item]

CUHK Distinguished Lecture Wednesday 17^th June 2020

Professor Sethu Vijayakumar presented a distinguished lecture online ‘From Automation to Autonomy: Advances in Machine Learning driving Next-gen Robotics’. The talk briefly introduced powerful machine learning technologies ranging from robust multi-modal sensing, shared representations, scalable real-time learning and adaption and compliant actuation that are enabling us to reap the benefits of increased autonomy while still feeling securely in control.

The CogX Global Leadership Summit and Festival of AI & Breakthrough Technology Monday 8^th June 2020

Professor Sethu Vijayakumar led a panel discussion on ‘Collaborative Robotics and Autonomous Systems’ with the theme ‘How do we get the next 10 years right?’ [News item]

ICRA 2020 International Conference on Robotics and Automation 4^th June 2020

Supported by the Alan Turing Institute. Workshop Information

Edinburgh International Science Festival 11-15 April 2020 *Cancelled*

Professor Sethu Vijayakumar to represent Edinburgh Centre for Robotics and the Alan Turing Institute at a series of events including:

Talos Humanoid Robotics Showcase
The Robotarium
RSE After hours

Upcoming events

Cambridge Consultants Roundtable Tokyo 14^th October 2020

Professor Sethu Vijayakumar invited to take part in the Collaborations in AI roundtable workshop.

IEEE UK RAS UK & I Conference Wednesday 27^th January 2021 *Postponed*

The annual IEEE UK & I RAS conference will be held at the University of Edinburgh hosted by Professor Sethu Vijayakumar. The aim of the conference is to improve the communication between members, researchers, students and industrial professionals who are interested in the activities of RAS in research, development and education. This event is aimed at the next generation of roboticist and combines a poster session with a series of talks by leading UK researchers.

UK-Italy Robotics and AI Collaboration Workshop February or March 2021

Edinburgh International Science Festival April 2021

There will be a Talos and Nextage demonstration carried out during the2021 International Festival.

Other activity

Professor Nick Hawes joined the AI (Robotics) programme as Group Leader.

Professor Sethu Vijayakumar elected fellow of the ELLIS network (The European Lab for Learning and Intelligent Systems) and will be involved with the program ‘Ellis Robot Learning: Closing the Reality Gap!’ [News item]

Number of current active research projects: 4

Number of completed research projects: 0

Number of projects in the pipeline: 0

Research Project 1 – Safe AI for surgical assistance

Project Duration: 01/01/2019 – 31/12/2020

PI – Dr Subramanian Ramamoorthy, University of Edinburgh

Co-I – Dr Paul Brennan, University of Edinburgh

Project website

Notable activity/outcomes in research progress

Development and robotic demonstration of methods for learning with requirements associated with safety and explainability.

Demonstration of lab scale prototype of these methods in tasks including soft material manipulation, ultrasound examination etc.

Ongoing development of software libraries associated with this methodology.

Status of progress against goals, any blockers to progress

Our primary milestones include: (1) software tools for model specification and synthesis, (2) demonstrators acting as proof of principle for surgical robotics, (3) Applications for larger scale projects

We are on track to achieving all of these milestones:

We have implemented lab scale demonstrators, which have now allowed us to design experiments that come closer to use cases of interest to surgical robotics (ongoing work with key deadlines in spring 2021).

We have a software repository in GitHub which we will eventually package in a form that can be distributed more widely. Not included in this open repository yet is an internal development of an interactive tool for learning specifications in an executable representation.

We have already initiated several follow-on projects, listed below. This is in addition to a specific targeted application in preparation, aimed at the surgical robotics use cases.

There are no specific blockers at the moment, but we do note:

Our laboratory experiments were significantly impacted by Covid lockdowns, due to which we requested an extension of the project end data.

While we are well on track to conducting proof of concepts in our lab, there are still some unknowns about how close to the medical use cases we can get by the end of this project. This is more a question of how much impact we can achieve rather than one of whether we can do what we have set out to achieve. We are actively working on this front.

Papers published/submitted

Under Review. Submitted

M. Burke, K. Lu, D. Angelov, A. Straižys, C. Innes, K. Subr, S. Ramamoorthy, Learning robotic ultrasound scanning using probabilistic temporal ranking. https://arxiv.org/abs/2002.01240, https://sites.google.com/view/ultrasound-scanner

T. Davchev, K. S. Luck, M. Burke, F. Meier, S. Schaal, S. Ramamoorthy, Residual learning from demonstration. https://arxiv.org/abs/2008.07682, https://sites.google.com/view/rlfd

M. Burke, K. Subr, S. Ramamoorthy, Action sequencing using visual permutations. https://arxiv.org/abs/2008.01156

Y. Hristov, S. Ramamoorthy, Learning from demonstration with weakly supervised disentanglement. https://arxiv.org/abs/2006.09107, https://sites.google.com/view/weak-label-lfd

P. Ardón, È. Pairet, K. Lohan, S. Ramamoorthy, R. Petrick, Affordances in robotic tasks -- a survey. https://arxiv.org/abs/2004.07400

Published in peer reviewed conference proceedings/journals

E. Kahembwe, S. Ramamoorthy, Lower dimensional kernels for video discriminators, Neural Networks Journal, Special Issue on Deep Neural Network Representation and Generative Adversarial Learning, 2020. https://arxiv.org/abs/1912.08860

D. Angelov, Y. Hristov, M. Burke, S. Ramamoorthy, Composing diverse policies for temporally extended tasks, IEEE Robotics and Automation Letters, Vol 5(2): 2658-2665, 2020. https://arxiv.org/abs/1907.08199,

https://www.youtube.com/watch?v=b2WhYp2aNC8&feature=youtu.be

M. Asenov, M. Burke, D. Angelov, T. Davchev, K. Subr, S. Ramamoorthy, Vid2Param: Modelling of dynamics parameters from video, IEEE Robotics and Automation Letters, Vol 5(2): 414-421, 2020. To be presented at the IEEE International Conference on Robotics and Automation, 2020. https://arxiv.org/abs/1907.06422, https://www.youtube.com/watch?v=JqtJYrj0mHs

C. Innes, S. Ramamoorthy, Elaborating on learned demonstrations with temporal logic specifications, Robotics: Science and Systems (R:SS), 2020. https://arxiv.org/abs/2002.00784, https://sites.google.com/view/ltl-dmp-rss-2020/

A. Straizys, M. Burke, S. Ramamoorthy, Surfing on an uncertain edge: Precision cutting of soft tissue using torque-based medium classification, In Proc. IEEE International Conference on Robotics and Automation (ICRA), 2020. https://arxiv.org/abs/1909.07247, https://www.youtube.com/watch?v=FNony5cUujA

Software developed/code released/New datasets created/accessed

GitHub (Edinburgh university dual arm software repository - built for exploration into learning methods for safer surgery through robot assistance).

Current functionality includes: motion planning of dual arm setup, 3D mouse control with Spacenav, User defined contour point following, Kinesthetic demonstration, autonomous ultrasound scanning.

External engagement, influence, impact (academic/ industry/ govt./ public/ international)

Cabaret of Dangerous Ideas (CoDI), Edinburgh Festival, 15 July 2020.

ISOLATED talks from home, June 2020.

XpoNorth Shorts EP 17 with Jessica Fox, June 2020.

Leveraged funding/support

Leading the UKRI Research Node on Trustworthy Autonomous Systems Governance and Regulation. This is a £3.3 million investment, as part of the TAS programme which includes a Hub and six other nodes.

Building directly on the Turing project, submitted an outline stage application to the second EPSRC Transformative Healthcare Technologies call.

Synergistic with the Turing project, receiving an unrestricted gift from Microsoft Research, which is to support the group's work on neural network based simulators of dynamical systems.

CoI in UoE - L&G Advanced Care Research Centre: New Technologies of Care, 2020 - 25. (£20 Million investment).

CoI in “A Cyber-Physical System for Unified Diagnosis and Treatment of Lung Diseases”, UKRI Future Leaders Fellowship awarded to M. Khadem, MRC Reference MR/T023252/1, 2020 - 24. (£1 Million).

PI, Metrology Fellowship: Trustworthy Control Synthesis, National Physical Laboratory, 2020-21. (£163K)

Research Project 2 – Intuitive human-robot interaction in work environments

Project Duration: 01/08/2019 – 31/07/2020 – EXTENDED UNTIL 31/12/2020

PI – Professor Kaspar Althoefer, Queen Mary University of London

Co-I – Professor Patrick Healey, Queen Mary University of London, Ildar Farkhatdinov Queen Mary University of London (QMUL), Julian Hough Queen Mary University of London (QMUL), Lorenzo Jamone, Queen Mary University of London (QMUL)

Project Website

Notable activity/outcomes in research progress

Journal and conference papers summarising the projects advancements prepared/published.

ICRA Workshop on human-robot tool handover successfully run, facilitating the exchange of thought between international experts.

Status of progress against goals, any blockers to progress

One of the main foci of this project is the preparation of grant proposals. Progress has been made over the summer. Three grant applications have been drafted and are in a promising state. However, because of the on-going COVID19 pandemic progress is slowed down. We expect to complete the grant writing in the first half of 2021. We have applied for an extension to accommodate this delay.

Progress was also delayed because the originally hired research assistant left for another position. The subsequently employed research assistant also left for another job after a short period. We are now in the process of finding a replacement for the open position.

Work on experimental studies of human-robot handover made initially very good progress; however, this is also delayed now mainly because of research assistants leaving and the COVID19 pandemic.

Conference/workshop talks presented

We organised the ICRA 2020 Workshop on Human-Robot handovers, 4th June 2020; this workshop was co-sponsored by the Alan Turing Institute. The workshop was very successful and allowed us to exchange thoughts and ideas with experts in the field. The work presented by the international colleagues was informative for us and gave us new ideas for the planned grant applications.

Papers submitted or published

F. Palermo, J. Konstantinova, K. Althoefer, S. Poslad and I. Farkhatdinov, "Implementing Tactile and Proximity Sensing for Crack Detection," 2020 IEEE International Conference on Robotics and Automation (ICRA), Paris, France, 2020, pp. 632-637. (https://qmro.qmul.ac.uk/xmlui/bitstream/handle/123456789/65151/Farkhatdinov%20Iimplementing%20Tactile%20and%202020%20Accepted.pdf?sequence=2)

F. Palermo, J. Konstantinova, K. Althoefer, S. Poslad and I. Farkhatdinov, Automatic Fracture Characterization Using Tactile and Proximity Optical Sensing, Frontiers in Robotics (conditionally accepted).

Förster, F., Althoefer, K., Attribution of Autonomy and its Role in Robotic Language Acquisition. AI & Society (accepted) 

New datasets created/accessed

Tactile sensing dataset for mechanical fractures characterisation, https://figshare.com/s/14deb00d874400e34d6714deb00d874400e34d67

Leveraged funding/support

EPSRC New Investigator Award, EP/T027746/1, Automatic Posture and Balance Support for Supernumerary Robotic Limbs (PI: Farkhatdinov).

EPSRC New Investigator Award for machine learning human-inspired robotic manipulation (Jamone, EP/S00453X/1, £325K).

Patents

Patent application on novel eversion-type robot filed (Althoefer, 28/08/2020)

Research Project 3 – Flexible autonomy for swarm robotics

Project Duration: 01/08/2019 – 31/07/2020

PI – Professor Sarvapali Ramchurn, University of Southampton

Co-I – Dr Danesh Tarapore, University of Southampton

Project Website

Notable activity/outcomes in research progress

Progress on new machine learning approach for human-swarm teaming.

Status of progress against goals, any blockers to progress

Goals met (papers submitted, platform nearly finalised).

Papers submitted or published

Paper accepted in ANT 2020, proceedings will be published in the Springer LNCS series.

External engagement, influence, impact (Academic/industry/govt./public/international)

Collaboration with Dresden University of Applied Sciences, Germany.

Collaboration with DSTL for further development.

Leveraged funding/support

Grant confirmed for enhancement project (5 Months).

Awards / Recognition

Won the bid to run the UKRI Trustworthy Autonomous Systems Hub.

Research Project 4 – Model criticism in Multi-Agent systems

Project Duration: 01/01/2020 – 31/12/2021

PI – Dr Stefano V. Albrecht

Project Website

Status of progress against goals, any blockers to progress

Turing project is on track. We submitted a paper to IAAI for the security work, and are preparing two papers for submission to ICRA for the robotics work (see below).

External engagement, influence, impact (Academic/industry/government/public/international)

After the Turing MAS Symposium in Feb, we applied to open a new MAS special interest group at the Turing. The application was successful and we're now in the process of setting this up.

Papers submitted or published

- Paper accepted at NeurIPS (https://arxiv.org/abs/2006.07169) - Submitted paper to IAAI (https://aaai.org/Conferences/AAAI-21/iaai-21-call/) - Submitted paper to AAAI (https://aaai.org/Conferences/AAAI-21/aaai21call/) - Submitted paper to ICLR (https://iclr.cc/Conferences/2021/Dates) - Submitted paper to AAMAS (https://aamas2021.soton.ac.uk/) - Will submit two papers to ICRA, due end of Oct 2020 (http://www.icra2021.org/)

Software/code/tools/methods developed/released

Released code for Multi-Agent test environments, and a full implementation of our protocol for quantum-secure authentication and key agreement (submitted to IAAI): GitHub

Articles/blogs/media/press releases (include links)

Blog coverage of Turing MAS Symposium

Leveraged funding/support

Quantum-secure authentication and key agreement

Principal Investigator - Dr Stefano Albrecht

Application submitted to Leverhulme but not successful (no feedback or reviews); now preparing to resubmit to EPSRC.

Applied to Innovate UK "Sustainable Innovation Fund" for industry project funding with bike logistics company "Pedal Me" in London. No outcome yet.

Google donated $10k to our lab for compute credits on their cloud platform.

Other research projects/activity

Upcoming

Two Postdoctoral Researchers and one Lab Manager position co-badged with Alan Turing to be advertised before the end of 2020.

Papers submitted / published

'Automatic Gait Pattern Selection for Legged Robots'. Authors: Jiayi Wang, Iordanis Chatzinikolaidis, Carlos Mastalli, Wouter Wolfslag, Guiyang Xin, Steve Tonneau and Sethu Vijayakumar. Proc. IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS 2020), Las Vegas, USA (2020)

‘Crocoddyl: An Efficient and Versatile Framework for Multi-Contact Optimal Control’. Authors: Carlos Mastalli, Rohan Budhiraja, Wolfgang Merkt, Guilhem Saurel, Bilal Hammoud, Maximilien Naveau, Justin Carpentier, Ludovic Righetti, Sethu Vijayakumar and Nicolas Mansard. Proc. IEEE International Conference on Robotics and Automation (ICRA 2020), Paris, France (2020)

Programme Updates - January 2020 to March 2020

Highlights

UK Multi-Agent Systems Symposium (UK-MAS) held in London on 24 February 2020.

Strengths

UK Multi-Agent Systems Symposium (UK-MAS) held in London on 24 February 2020.

Opportunities

UK-Italy Robotics and AI Collaboration Workshop to be held later in 2020.

Involvement with the Japan AI Moonshot projects.

Challenges and risks

Number of events postponed or cancelled due to the international spread of the coronavirus (COVID-19).

Upcoming plans/priorities

Seek support from BEIS, OAI, UKRI and on Japanese side (METI, MEXT) on broader project proposal outcomes from the UK-Japan workshop held in September.

General programme updates

Past Events

UK Multi-Agent Systems Symposium (UK-MAS) – 24 February 2020, London

Lead organiser Dr Stefano V. Albrecht held the first UK Multi-Agent Systems Symposium in London with the aim to explore the MAS research landscape in the UK by bringing together UK based research labs at Universities and in Industry who have a significant focus on MAS research.

The symposium, which was fully subscribed, included 20 spotlight presentations along with presentations from Edward Hughes and Yoram Bachrach from Deepmind, Sam Devlin from Microsoft Research and Subramanian Ramamoorthy from FiveAI.

Presentations were followed by a discussion and an opportunity for a networking session.

Upcoming events

UK-Italy Robotics and AI Collaboration Workshop led by Professor Angelo Cangelosi. *Rescheduled date to be confirmed*

Aim to explore collaboration opportunities between the two countries. The Italian organisations initially involved in this activity are CINI-AIIS (Consorzio Interuniversitario Nazionale per Informatica - Laboratorio Nazionale di Artificial Intelligence and Intelligent Systems), Artes 4.0 (Centro di Competenza ad Alta Specializzazione in Industria 4.0 e Robotica) and IIT (Istituto Italiano di Tecnologia). The workshop will bring together leading researchers in AI and robotics from Italy and the UK.

AI UK 24-25 March 2020 **POSTPONED**

Two day event hosted by the Alan Turing Institute showcasing the best of UK Academic work in AI and Machine Learning. Professor Sethu Vijayakumar due to present on Creating Controls: Robotics and automated systems

IEEE UK RAS UK & I Conference Wednesday 27^th January 2021

This event is aimed at the next generation of roboticist and combines a poster session with a series of talks by leading UK researchers.

Meetings/presentations/engagement

Edinburgh International Science Festival 11-15 April 2020 (CANCELLED)

Professor Sethu Vijayakumar to represent Edinburgh Centre for Robotics and the Alan Turing Institute at a series of events including

Talos Humanoid Robotics Showcase

The Robotarium

RSE After Hours

ICRA 2020 International Conference on Robotics and Automation 4^th June 2020, Paris supported by the Alan Turing Institute (CANCELLED)

Number of current active research projects: 4

Number of completed research projects: 0

Number of projects in the pipeline: 4

Research Project 1 – Safe AI for surgical assistance

Project Duration: 01/01/2019 – 31/12/2020

PI – Dr Subramanian Ramamoorthy, University of Edinburgh

Co-I – Dr Paul Brennan, University of Edinburgh

Project website

Funded by – Turing Fellow projects

Notable activity/outcomes in research progress

The major outcome of this quarter is to consolidate our work into two major use cases that line up better for subsequent external engagement and larger grants. These include:

Multi-modal system identification and model learning for intra-operative guidance.

Incremental interactive elaboration of safety envelopes and task specifications

Together these form new functionality that can be incorporated into next generation medical robotics systems.

Status of progress against goals, any blockers to progress

Goal 1 – Software tools for model learning and controller synthesis (partially met, work continues to develop)

Goal 2 – Demonstrator of proof of principle – autonomous robotic surgery (partially met, ongoing work)

Goal 3 – Applications for larger scale projects building on goal 2 (partially met, see below. Two grant applications in prep for submission in mid-year 2020)

Conference/workshop talks presented

International Workshop on Intelligent Robot Teammates for Complex Missions in Unstructured Environments – Joint Advanced Technology Centre, Indian Institute of Technology, Delhi, 6^th March 2020.

UK Multi-Agent Systems Workshop – Alan Turing Institute, London 24 February 2020.

Papers published/submitted

Published:

P. Ardón, È. Pairet, R. Petrick, S. Ramamoorthy, and K. Lohan, Learning grasp affordance reasoning through semantic relations, IEEE Robotics and Automation Letters 4(4): 4571-4578, 2019. Presented at the IEEE International Conference on Intelligent Robots and Systems, 2019. https://ieeexplore.ieee.org/document/8790777

Submitted:

P. Ardón, È. Pairet, K. Lohan, S. Ramamoorthy, R. Petrick, Affordances in Robotic Tasks -- A Survey. https://arxiv.org/abs/2004.07400

M. Burke, K. Lu, D. Angelov, A. Straižys, C. Innes, K. Subr, S. Ramamoorthy, Learning robotic ultrasound scanning using probabilistic temporal ranking. https://sites.google.com/view/ultrasound-scanner

C. Innes, S. Ramamoorthy, Elaborating on learned demonstrations with temporal logic specifications. https://arxiv.org/abs/2002.00784

E. Kahembwe, S. Ramamoorthy, Lower dimensional kernels for video discriminators. https://arxiv.org/abs/1912.08860

Software developed/code released/New datasets created/accessed

We maintain this GitHub repository which summaries project outcomes:

New datasets created / accessed

Through synergistic Cancer Research UK funded project, we have started to work with a pCLE imaging of breast cancer dataset (created at Imperial College).

External engagement, influence, impact (academic/ industry/ govt./ public/ international)

Malvern Festival of Ideas (6-8 March 2020)

1^st Scottish Online Seminar on Robots in Healthcare, 27^th April 2020

Articles/blogs/media/press releases

Interview for XpoNorth Digital Shorts, 8^th May 2020

Leveraged funding/support

Co-I, A Cyber-Physical System for Unified Diagnosis and Treatment of Lung Diseases, UKRI Future Leaders

Fellowship - Mohsen Khadem, MRC Reference MR/T023252/1, 2020 - 24. (£1M)

Co-I, MAMMOBOT: A flexible robot for early breast cancer diagnosis, Cancer Research UK, 2019 - 21.

(£100K. We reported this earlier but the official start date was 1 Jan 2020, it is now ongoing)

This work feeds into my involvement in the following applications: (1) UKRI TAS Hub, Turing Institute led application, (2) UKRI TAS Node on Verifiability, Co-I, (3) UKRI TAS

Patents

We are beginning to work on a patent for a device created as part of this project (further details in next quarterly report)

Research Project 2 – Intuitive human-robot interaction in work environments

Project Duration: 01/08/2019 – 31/07/2020 – EXTENDED UNTIL 31/12/2020

PI – Professor Kaspar Althoefer, Queen Mary University of London

Co-I – Professor Patrick Healey, Queen Mary University of London

Project website

Notable activity/outcomes in research progress

Following the previous reporting period, we continued to work on proposals with different project partners. Three proposals are currently being actively pursued, one targeting surgical theatres, a second one targeting the regular manufacturing industry, and the third one targeting tasks revolving around nuclear decommissioning. All three different proposals target the EPSRC as funder, and a potential fourth proposal may target the Wellcome. At this moment most advanced is the proposal draft targeting surgical theatres with a preliminary intention to submit the application in June 2020. For the nuclear proposal an additional partner, Birmingham University, was added to the consortium. An abstract has been drafted to be reviewed by one or two contact persons closer to or directly involved with the nuclear industry to which project partners have links to. By obtaining feedback from persons from the nuclear industry we hope to be able to address the industry’s actual needs better and thereby increase the likelihood of the proposal being funded.

Status of progress against goals, any blockers to progress

The overall goals of the project remain identical to those listed in the report of the previous quarterly report. These consist of the preparation of 3+1 grant applications jointly with partners across the UK. The `+1’ is ought to indicate the aforementioned potential application to the Wellcome Trust, which may happen once the three EPSRC proposals will have been submitted.

Intermediate Goals and Progress against these:

Preparation of joint grant application targeting surgical setting (achieved: initial draft of grant application for surgical strand sketching out the temporal structure of the project including work packages and an iterative experimental design)

Preparation of joint grant application targeting industrial setting (achieved: progressed talks with partners from UCL, early incomplete version of grant application)

Preparation of joint grant application targeting nuclear setting (achieved: held several teleconferences with prospective partner, drafted proposal abstract for obtaining feedback from nuclear industry contacts)

Preparation of observational study on human-human handovers in surgical theatres (achieved: initiated NHS ethics application)

Presented at workshop on Test Methods and Metric for Effective HRI in Real World Human-Robot Teams (see above).

Continued planning of ICRA 2020 workshop on human-robot handovers (exceeding project objectives).

Conference/workshop talks presented

ICRA Workshop Organisation

Due to the Covid-19 pandemic it was unclear whether the ICRA 2020 conference for which we had planned a workshop would take place beginning of June and if yes, in what format. Only this week we were informed that the conference would take place virtually, and we are currently deciding on the concrete format of the workshop and the online platform to be utilised. We currently assume that the workshop will take place, but need to consult with the invited speakers in order to ensure that they are still willing to contribute to the workshop in its online format.

Abstracts for presentation for two workshops on the ACM/IEEE International Conference for Human-Robot Interaction (HRI 2020) have been accepted by the respective workshop organisers. Due to Covid-19, HRI 2020 was unfortunately cancelled and with it one of the workshops. However, the workshop “Test Methods and Metrics for Effective HRI in Real World Human-Robot Teams” did take place in an online format, and we gave a presentation on the prospects of measuring motor resonance in real-world HRI scenarios. We intend to submit an article in the forthcoming special issue “Test methods for human-robot teaming performance evaluations” linked to this workshop (journal: IEEE Transactions on Human-Robot Interaction).

Papers submitted or published

A position paper related to but more general than handover tasks was submitted to the IEEE International Conference on Robot and Human Interactive Communication (Ro-MAN 2020) and is currently under review.

A journal paper on perceived autonomy building up on previous work the research associate working under the current grant, was submitted to a special issue on “Drones, robots and perceived autonomy” of the journal “AI & Society”. The paper is currently under review.

Research Project 3 – Flexible autonomy for swarm robotics

Project Duration: 01/08/2019 – 31/07/2020

PI – Professor Sarvapali Ramchurn, University of Southampton

Co-I – Dr Danesh Tarapore, University of Southampton

Project website

Notable activity/outcomes in research progress

Testbed development to evaluate swarming algorithms and to test state estimation/prediction.

Status of progress against goals, any blockers to progress

Progress as expected.

Extension requested to cater for later starting date.

Conference/workshop talks presented

Presentation at Southampton Turing briefing.

Papers submitted or published

Paper submitted to EUMAS and being revised for DARS.

Software/code/tools/methods developed/released

Software toolkit developed and trial being setup with Dstl.

New datasets created/accessed

Test cases developed to include environments with occlusion and communication dropouts.

External engagement, influence, impact (Academic/industry/govt./public/international)

Engagement with Dstl, Thales and BAE

Leveraged funding/support

Contribution from Dstl for a CDT studentship (£100k)

Research Project 4 – Model criticism in Multi-Agent systems

Project Duration: 01/01/2020 – 31/12/2021

PI – Dr Stefano V. Albrecht

Project website

Notable activity/outcomes in research progress

None to report yet.

Status of progress against goals, any blockers to progress

Progress is good so far. We are preparing a submission to NIPS which is a top ML conference and due end of May. This is for our cyber security work as detailed in proposal. Two interns were also hired to bolster our work in autonomous driving over the summer (4 months each). For that work we are targeting a robotics conference (ICRA) which is due in September.

External engagement, influence, impact (Academic/industry/government/public/international)

Hugely successful UK Multi-Agent Systems Symposium (UK-MAS) held on the 24 February 2020, London.

Further information on the event can be found earlier in the report under past events.

Articles/blogs/media/press releases (include links)

Kate Highnam, Imperial College London wrote a blog on ‘The Future of Multi-Agent Systems Research’ and Joseph Early, The Alan Turing Institute wrote a blog on ‘What’s hot in Multi-Agent Systems?’

Leveraged funding/support

£15k received from the Royal Society and UKRI towards the hiring of two interns.

Other research projects/activity Ongoing

Dr Cristian Rauch and Dr Sanghyun Kim appointed to Post-doctoral positions with ORCA, co-badged with Alan Turing.

Dr Josiah Hannah appointed to work with Dr Stefano Albrecht on the project “Model criticism in Multi-Agent systems”.

Papers submitted / published

Carlos Mastalli, Rohan Budhiraja, Wolfgang Xaver Merkt, Guilhem Saurel, Bilal Hammoud, Maximilien Naveau, Justin Carpentier, Ludovic Righetti, Sethu Vijayakumar and Nicolas Mansard, Crocoddyl: An Efficient and Versatile Framework for Multi-Contact Optimal Control, Proc. IEEE International Conference on Robotics and Automation (ICRA 2020), Paris, France (2020).

Programme Updates - October 2019 to December 2019

Highlights

UK-Japan project awarded funding for joint project for UKRI-JST Joint call of Artificial Intelligence and Society
New CUHK-SZ AIRS institute funding (CNY 3,000,000) for robotics and AI activities based on a dumb-bell model between CUHK-SZ and UoE/Turing Robotics Hub

Strengths

UK-Japan project awarded funding for joint project for UKRI-JST Joint call of Artificial Intelligence and Society

Opportunities

UK-Italy Robotics and AI Collaboration Workshop
Involvement with the newly announced Japan AI moonshot projects

Challenges and risks

We need to engage better and faster with BEIS/OAI to leverage opportunities arising from the UK-Japan engagement

Upcoming plans/priorities

Seek support from BEIS, OAI, UKRI and on Japanese side (METI, MEXT) on broader project proposal outcomes from the UK-Japan workshop held in September.

General programme updates

Upcoming events

UK Multi-Agent Systems Symposium (UK-MAS) 24 February 2020, London. Led by AI (Robotics) Group Leader Dr. Stefano V. Albrecht.

The goal of the symposium is to bring together UK-based research labs at Universities and Industry who have a focus on MAS research, to explore the MAS research landscape in the UK. There will be space for talks by people from academia and industry as well as discussion sessions and networking opportunities. Outcomes include a virtual map with the locations and research focus of the various UK-based labs, as well as a shared mailing list for future collaborations.

UK-Italy Robotics and AI Collaboration Workshop 18-19 March 2020, Rome. Led by Professor Angelo Cangelosi.

Meetings/presentations/engagement

Professor Sethu Vijayakumar presented at the IROS 2019 conference in Macau in November 2019.

Dr Stefano V. Albrecht presented his work with FiveAI on autonomous driving at the Royal Society Industry Networking event in November 2019.

Professor Sethu Vijayakumar to present at the ‘Hello Robot’ exhibition at V&A Dundee in January 2020 to discuss the work carried out at Edinburgh Centre for Robotics and The Alan Turing Institute [News item]

Professor Sethu Vijayakumar to give a research talk at Advanced Telecommunication Research (ATR) Labs, Kyoto regarding Turing Activities on 10 February 2020.

Professor Sethu Vijayakumar to give a keynote speech at AIST-AIRC Annual Symposium in Tokyo on 19 February and a talk at the new SONY AI labs, Tokyo, on Robot Control and Multi Contact Optimisation on the 18 February 2020.

Other activity

New UoE/Turing collaboration and setup of joint Artificial Intelligence and Robotics for Society (AIRS) institute under the umbrella of Chinese University of Hong Kong -Shenzhen Campus in collaboration with Stanford, U.Penn, Michigan etc. -- Professor Sethu Vijayakumar will lead the UK effort and is negotiating an MoU with Turing, with a view to provide funding for hiring 3 postdoctoral fellows (total funding of around RMB 3,000,000) [Project page]

Discussion taking place with Tencent’s Robotics X division regarding support for research on Multi-contact Robot Control and Planning with Holistic Multimodal (force, tactile, visual) Perception. A ‘gift’ is currently being negotiated.
Students from Dr Stefano V. Albrecht’s group placed 4^th in the recent Huawei DriveML competition. More than 200 teams were registered and around 25 teams made it to the final stage. [Competition page]

Turing’s Public Policy Programme ethics fellow has been awarded funding for a joint project for UKRI-JST Joint Call of Artificial Intelligence and Society

Title: PATH-AI: Mapping an Intercultural Path to Privacy, Agency and Trust in Human-AI Ecosystems

Collaborating Institutions: RIKEN AIP (Hiroshi Nakagawa, PI) & Alan Turing Institute (Dr David Leslie, PI).

Timeframe: 15 January 2020 – 14 February 2023

Funding amount: £484,657.00 on UK side

AIJ survey (https://www.sciencedirect.com/science/article/abs/pii/S0004370218300249) by Dr Stefano V. Albrecht has been named in the list of most cited papers in AIJ since 2016.

https://www.journals.elsevier.com/artificial-intelligence/most-cited-articles

Number of current active research projects: 3

Number of completed research projects: 0

Number of projects in the pipeline: 4

Research Project 1 – Safe AI for surgical assistance

Project Duration: 01/01/2019 – 31/12/2020

PI – Dr Subramanian Ramamoorthy, University of Edinburgh

Co-I – Dr Paul Brennan, University of Edinburgh

Project website

Funded by – Turing Fellow projects

Status of progress against goals, any blockers to progress

Progress is on track.

Papers published/submitted

Submitted:

E. Kahembwe, S. Ramamoorthy, Lower dimensional kernels for video discriminators. Subm. to Neural Networks journal. https://arxiv.org/abs/1912.08860

A. Straizys, M. Burke, S. Ramamoorthy, Surfing on an uncertain edge: Precision cutting of soft tissue using torque-based medium classification. Under review at ICRA 2020.

https://arxiv.org/abs/1909.07247

Published:

M. Burke, Y. Hristov, S. Ramamoorthy, Hybrid system identification using switching density networks, Conference on Robot Learning (CoRL), 2019. https://arxiv.org/abs/1907.04360

Y. Hristov, D. Angelov, A.Lascarides, M. Burke, S. Ramamoorthy, Disentangled Relational Representations for Explaining and Learning from Demonstration, Conference on Robot Learning (CoRL), 2019. https://arxiv.org/abs/1907.13627

D. Angelov, Y. Hristov, M. Burke, S. Ramamoorthy, Composing diverse policies for temporally extended tasks. Accepted to IEEE Robotics and Automation Letters, 2020.

https://arxiv.org/abs/1907.08199

Software developed/code released/New datasets created/accessed

Github repository updated to include major project threads

In conjunction with CRUK project below, we are setting up access to a breast cancer tissue bank and associated datasets.

External engagement, influence, impact (academic/ industry/ govt./ public/ international)

Demonstration to Douglas Ross, MP, Parliamentary Under Secretary of State for Scotland.

Expert forum on ‘AI in minimally invasive surgery and treatment’, organised by UK Science and Innovation Network and the German Federal Ministry for Economic Affairs and Energy, British Embassy, Berlin on 2-3 December 2019.

Leveraged funding/support

Cancer Research UK funded project Mammobot started in January 2020.

Awards/recognition

CoRL 2019 Best Paper Award Runner-up https://www.robot-learning.org

Research Project 2 – Intuitive human-robot interaction in work environments

Project Duration: 01/08/2019 – 31/07/2020

PI – Professor Kaspar Althoefer, Queen Mary University of London

Co-I – Professor Patrick Healey, Queen Mary University of London

Project website

Notable activity/outcomes in research progress

Continuation of Work on Project proposals

In the reporting period we continued to work on proposals with project partners from the previous quarter. The plan to submit at least three grant proposals to the EPSRC and the Wellcome Trust during the grant period remains unchanged, covering human-robot join action tasks in nuclear decommissioning, assistance in the surgical theatre, and the general industry.

Informal of Observations of Surgical Operations

In preparation for a planned observational study several members of the project team attended surgical procedures in the Royal London Hospital in order to get a realistic impression of processes and requirement. These observations were made possible through the support of the respective clinical surgeons who invited us to attend. The observed procedures fell into two categories, a colorectal surgery and a brain surgery, which are performed in general and neurological theatres respectively.

Planning of Observational Study on Human-Human Interactions in the Surgical Theatre

The observations mentioned in the previous paragraph provided the basis for the planning of a future study on interactions between various team members during surgical procedures. The main focus will be on interactions between the surgeons, scrub nurses, and circulating nurses. Given the interactional complexity of the tasks the study focusses on the creation of a reference data set of real-world interactions in the surgical theatre which is ought to provide the basis for future robotic work. We intend to derive a set of requirements from this data set that specifies targets for the various robotic and AI components that would needed for a robot to engage in handover tasks in the surgical setting, with components ranging from planners and dialogue systems to dextrous manipulators. Currently we intend to cooperate with researchers from UCL and Kings College in order to extend the number of hospitals targeted for observation, and potentially extend the types of surgical procedures and theatres beyond the two aforementioned ones.

Conference/workshop talks presented

Within the reporting period we applied successfully to hold an interdisciplinary workshop on human-robot handovers at the International Conference on Robotics and Automation (ICRA 2020). The workshop will take place either on the 31^st of May or on 4^th of June in Paris. Eleven invited speakers could be won to present their work on human-robot (or robot-human) handovers and related human-robot joint actions. The speakers hale from sociology (Christian Heath, Kings College), over cognitive science (Günther Knoblich, Central European University) to robotics (Aude Billard, École Polytechnique Fédérale de Lausanne), and most are world-leading experts in this topic, but with differing methodologies compliant with their respective scientific field. We hope to gain greater insights into the issues associated with handover operations by bringing together experts from different fields, instigate discussions and hopefully trigger future collaborations.

We also managed to obtain support for this workshop from three different IEEE Technical Committees (Cognitive Robotics, Human Movement Understanding, Human-Robot Interaction & Coordination).

We are currently in negotiations with ACM Transactions on Human-Robot Interaction about a special issue that may arise from the workshop.

Papers submitted or published

A position paper related but more general than handover tasks was submitted to the alt.HRI stream of the ACM/IEEE International Conference on Human-Robot Interaction (HRI 2020). Unfortunately the submission was not successful, but we intend to submit an extended version of the paper to IEEE International Conference on Robot and Human Interactive Communication (Ro-MAN 2020).

Status of progress against goals, any blockers to progress

Overall Goals

The overall goals remain identical to those listed in the report of the previous quarterly report, namely the preparation of three (or more) grant applications jointly with partners across the UK.

Intermediate Goals and Progress

Establishing contact with additional partners concerning the preparation of joint applications (achieved: in contact with Helge Wurdemann and Christian Heath; preparatory teleconference meetings held)
Preparation of observational study on human-human handovers in surgical theatres (achieved: informally attended surgical procedures to inform further planning of study and NHS ethics application. Extended partners in study to include Christian Heath and, potentially Helge Wurdemann.)
Attended workshop on impact measurement and management organised by Allia Future Business Centre (Cambridge) in order to improve impact-related statements in project applications .
Successfully applied for ICRA 2020 workshop on human-robot handovers (exceeding project (exceeding project objectives).

Research Project 3 – Flexible autonomy for swarm robotics

Project Start Date: August 2019

PI – Professor Sarvapali Ramchurn, University of Southampton

Co-I – Dr Danesh Tarapore, University of Southampton

Project Website

Notable activity/outcomes in research progress

New interaction modalities developed for human-swarm interaction. We have focused on two key aspects: (1) monitoring the state of the swarm (2) directing the swarm to carry out specific tasks in a predictable way when communication is limited and information is delayed. We have specifically focused on limiting the amount of information that an operator or multiple operators have to comprehend when carrying out both tasks.

We also held a workshop at The Alan Turing Institute on 1^st October 2019 – Event website

Status of progress against goals, any blockers to progress

Progressing on defining key use-cases for human-UAV teaming in disaster response.
Developed a simulator for human-swarm teaming.
Appointed an intern to support the work and potentially carry on to do a PhD.
Workshop held at the Turing helped expose the proposed ideas and develop some key directions for future research.

Conference/workshop talks presented

Presentation at IROS on “Plasticity in Collective Decision-Making for Robots” and a presentation at RO-MAN conference on “Responsible AI for Human-UAV Teaming”.

Papers submitted or published

Paper submitted to AI journal on Responsible AI.

Paper to be submitted to IJCAI (in Jan) on Human-Swarm teaming

Software developed/code released/New datasets created/accessed

A number of existing techniques for swarm coordination are being built into our simulator (to be open-sourced).

We are currently using synthetic datasets but are working with DSTL, Werobotics and Thales to acquire use-cases and datasets.

External engagement, influence, impact (Academic/industry/govt./public/international)

Contacts we are currently actively working with include:

DSTL AI Lab (Steve Meers)
BAE Systems (AIR)
Thales Autonomy

Advising RUSI and MOD.

Leveraged funding/support

£30K – In-kind contributions from DSTL

£200K – funds for two jointly funded PhD with DSTL and the MINDS CDT on Human-UAV Teaming

Two proposals in preparation directly building on this project.

Other research projects/activity

Ongoing

RA Josia Hannah appointed to work with Dr Stefano Albrecht on the project “Model criticism in multi-agent systems”. Start date January 2020.
Dr Stefano V. Albrecht and team currently working on five paper deadlines in January 2020.

Upcoming

Three Post-doctoral positions advertised through ORCA, co-badged with Alan Turing. Closing date 6^th January 2020.

Programme Updates - July 2019 - September 2019

Highlights

UK-Japan robotics and AI research collaboration workshop (17-19 September 2019)

Upcoming plans/priorities

Seek support from BEIS, OAI, UKRI and on Japanese side (METI, MEXT) on broader project proposal outcomes from the UK-Japan workshop

General programme updates

Past events

UK-Japan robotics and AI research collaboration workshop 17 – 18 September (Edinburgh) and 19 September (London). [News item]

This was the first formal collaboration between the UK and Japan in the field of robotics, AI and the ethical use of data as part of a wider UK government announcement, which was made earlier this year. Japanese institutions involved in the workshop included AIST, the National Institute of Advanced Industrial Science and Technology, NII, Japan’s National Institute of Informatics, and the RIKEN Centre for Advanced Intelligence Project.

Over 130 participants, including over 30 leading academic researchers from Japan and a further 30 industry representatives from UK and Japan, along with a number of UK academic researchers, attended the workshop. This was an important first step to create successful collaborations from the signed agreements and MoUs.

The workshop concluded with a networking event attended by all delegates within the Embassy of Japan, hosted by the Japanese Ambassador for the UK, Mr Koji Tsurouka. Guest speakers included Professor Sir Adrian Smith (Alan Turing) and Dr Rannia Leontaridi (BEIS).

Outcomes from the workshop will include researcher exchange placements, proposals for academic low-Technology Readiness Levels (TRL) projects, as well as industry-driven high-TRL research.

Over 25 project proposals were created during the three day workshop, which will bring together a number of new scientific collaborations between the UK and Japan.

Meetings/presentations/engagement

Professor Sethu Vijayakumar and two PhD Students visited IIT-Kanpur, India for a joint UK-India Project funded by UKIERI, focused on Autonomous Warehousing robots in July 2019.
Professor Sethu Vijayakumar was invited to deliver a keynote at the Nature AI and Robotics Conference in Shenzhen, hosted by Tencent in September 2019
Professor Sethu Vijayakumar is scheduled to present at the IROS 2019 Conference in Macau in November 2019

Other activity

New UoE/Turing collaboration and setup of joint Artificial Intelligence and Robotics for Society (AIRS) institute under the umbrella of Chinese University of Hong Kong -Shenzhen Campus in collaboration with Stanford, U.Penn, Michigan etc. -- Professor Sethu Vijayakumar will lead the UK effort and is negotiating an MoU with Turing, with a view to provide funding for hiring 3 postdoctoral fellows (total funding of around RMB 3,000,000) https://sse.cuhk.edu.cn/en/content/10865
Discussion taking place with Tencent’s Robotics X division regarding support for research on Multi-contact Robot Control and Planning with Holistic Multimodal (force, tactile, visual) Perception. A ‘gift’ is currently being negotiated.

Number of current active research projects: 4

Number of completed research projects: 0

Number of projects in the pipeline: 2

Research Project 1 – Safe AI for surgical assistance

01/01/2019 – 31/12/2020

PI – Dr Subramanian Ramamoorthy, University of Edinburgh

Co-I – Dr Paul Brennan, University of Edinburgh

Project website

Funded by – Turing Fellow projects

Notable activity/outcomes in research progress

Building new techniques for autonomous robotic surgery that will in due course make their way into new IP and potential medical products. At this stage we are still en-route to such outcomes.

Status of progress against goals, any blockers to progress

Goal 1 – Software tools for model learning and controller synthesis (Partially met)

Goal 2 – Demonstrator of proof of principle – autonomous robotic surgery (Partially met)

Goal 3 – Applications for larger scale projects building on goal 2 (Incomplete)

Conference/workshop talks presented

Robotics: Science and Systems

The Multi-disciplinary Conference on Reinforcement Learning and Decision Making

Papers published

M. Burke, S. Penkov, S. Ramamoorthy, From explanation to synthesis: Compositional program induction for learning from demonstration, Robotics: Science and Systems (R:SS), 2019. http://rss2019.informatik.uni-freiburg.de/papers/0077_FI.pdf

D. Angelov, Y. Hristov, S. Ramamoorthy, DynoPlan: Combining Motion Planning and Deep Neural Network based Controllers for Safe HRL, In Proc. The Multi-disciplinary Conference on Reinforcement Learning and Decision Making (RLDM), 2019.  http://rldm.org/papers/extendedabstracts.pdf

P. Ardón, È.Pairet, R. Petrick, S. Ramamoorthy, K. Lohan, Reasoning on grasp-action affordances, In Proc. Annual Conference Towards Autonomous Robotic Systems (TAROS), 2019. https://link.springer.com/chapter/10.1007/978-3-030-23807-0_1

*TAROS 2019 paper nominated as Best Paper Award Runner-up

International / industry engagements / meetings / presentations

Co-organiser for the Robust Artificial Intelligence for Neurobotics workshop; https://blogs.ed.ac.uk/rai-nr/

Group leader for discussions at the UK-Japan Robotics and AI collaboration workshop

Software developed/code released/New datasets created/accessed

Currently in the process of generating a new software code base and a dataset/competition based on our demonstrator, both to be released in 2020

Leveraged funding/support

Cancer Research £100,000

Edinburgh City Deal £250,000

Research Project 2 – Intuitive human-robot interaction in work environments

Date from- to

PI – Professor Kaspar Althoefer, Queen Mary University of London

Co-I – Professor Patrick Healey, Queen Mary University of London

Project website

The previous 3 months constitute the start period of the above-mentioned project. Within this time period. In August (month 2) of the project a Research Assistant started his role within the project and the following activities were initiated or accomplished.

Notable Activities

Drafting of Project proposals In the reporting period we contacted several potential future project partners. We are planning to submit at least three grant proposals to funders such as the EPSRC and the Wellcome Trust during the grant period and have started to work on two joint proposals. The underlying intention is to split the research agenda thematically along the three research areas mentioned in the project proposal – nuclear decommissioning operations, providing robotic assistance in the surgical theatre, and creating a robotic helper for more general industrial target scenarios.

Planning of Feasibility Study and Robotic Prototype

We started to design a study in order to obtain relevant data and technical expertise with respect to object and tool handovers.

The first planned study is a Human-Human Interaction (HHI) Study targeting the creation of a reference data set which is ought to inform subsequent robotic research. The study is intended to provide us with insights how human normally conduct handover operations under sub-optimal conditions such as the presence of noise, visual obstructions, or requiring non-standard body positions (kneeling, bent over, etc). These sub-optimal conditions are likely to be part of at least one of the three aforementioned proposals. We will therefore likely need more concrete data on how humans conduct handovers in such situations.

Apart from the aforementioned HHI study we intend to create a technical prototype involving a robot-arm in order to gain more technical expertise with respect to handover operations.

Workshop and Conference Talks

Within the reporting period we attended the `UK-Japan robotics and AI research collaboration workshop’ organised by the Turing institute and held in Edinburgh and London between the 17^th and 19^th of September. Kaspar Althoefer gave a talk on the groups’ work on the workshop and Frank Förster presented a poster on the forthcoming work of the present project.

The outcome of this workshop was an outline for future collaboration with Prof. Tetsuya Ogata from the Waseda University and Research Fellow at the Japanese research institution AIST. We intend to collaborate on a joint proposal once both the UK and Japanese governments have committed to more concrete funding plans for a joint programme.

The topic of the proposal outline focusses on object and tool handovers between robots and human.

Goals and Progress against Goals

Overall Goals The preparation of three (or more) grant applications jointly with partners across the UK. The proposed partners are largely the ones listed in the grant application, namely Helge Wurdemann (UCL), Rustam Stolkin (Birmingham), Angelo Cangelosi (Manchester), Christian Heath (KCL), and Manuel Giuliani (UWE).

Intermediate Goals and Progress against these

Establishing contact with proposed partners concerning the preparation of joint applications (achieved: in contact with Manuel Giuliani and Angelo Cangelosi; preparatory teleconference meetings held)

Preparation of drafts for grant applications (achieved: prepared two application drafts for projects on collaborative human-robot joints actions in an industrial context and human-robot handover operations in a nuclear decommissioning context, with Angelo Cangelosi and Manuel Giuliani respectively)

Feasibility study commenced (achieved: The study into tool handovers between robots and humans has started.)

Further collaboration beyond the project proposal (achieved: initial conversations with Prof. Tetsuya Ogata from the Waseda University held and outline proposal prepared.)

Attended at and contributed to Turing organized workshop (UK-Japan robotics and AI research collaboration) (exceeding project objective)

Research Project 3 – Flexible autonomy for swarm robotics

Date started August 2019

PI – Professor Sarvapali Ramchurn, University of Southampton

Co-I – Dr Danesh Tarapore, University of Southampton

Project website

Notable activity/outcomes in research progress

The project started in August 2019 with the appointment of a swarm robotics expert. To shape up the research, two meetings took place to engage with key partners, Thales and DSTL.

This was followed up with a workshop on the 1^st October at The Alan Turing Institute on Responsible human-machine teaming to further refine the interactional challenges to be addressed.

Status of progress against goals, any blockers to progress

A preliminary simulator has been designed and basic swarming algorithms are now being implemented. The goal is to develop some initial benchmarks and start to define challenges for human-swarm interaction.

Conference/workshop talks presented

Responsible human-machine teaming

Proposals for new projects

Agreement

Model criticism in multi-agent systems

Principal Investigator – Dr Stefano Albrecht

https://www.turing.ac.uk/research/research-projects/model-criticism-multi-agent-systems

Project due to start January 2020.

Negotiation

Mobile Collaborative Robots: Addressing Real World Challenges

Principal Investigator – Professor Sethu Vijayakumar

Co-Investigator – Dr Songyan Xin, Dr Lei Yan (affiliated to Alan Turing)

This project will address the challenges of developing a fully integrated mobile collaborative robot with critical functionalities such as base and arm synchronized planning algorithms, object-based environment representations and intuitive human-robot control interface.

Proposed dates: January 2020 – December 2021

Funding: Year 1 CNY 1,750,000

Year 2 CNY 950,000

Quantum-secure authentication and key agreement

Principal Investigator - Dr Stefano Albrecht

Application submitted to Leverhulme Trust (awaiting outcome)

Other research projects/activity

Ongoing

Three new Post-doctoral researchers hired through the ORCA / FAIRSPACE hub, co-badged with Alan Turing:

Joao Moura
Daniel Gordon
Songyan Xin
RA (Josia Hannah) appointed to work with Dr Stefano Albrecht on the project "Model criticism in multi-agent systems" - start date January 2020.

Upcoming

Further Post-doctoral positions to be advertised through ORCA/ FAIRSPACE hubs, co-badged with Alan Turing.

Year 1 - September 2018 to August 2019

1 September 2018: Professor Vijayakumar gave a keynote on Shared Autonomy at the World Congress of Robotics in Shenyang and talked about his new role as co-Director of the AI Programme, ATI. Follow up action on working with new Robotics Institute at North Eastern University (Thousand Talent Programme and ATI). WCR2018
3-7 September 2018: Professor Vijayakumar led the UK-Japan AI workshop in collaboration with the FCO and represented ATI to the Japanese Academics and Industry.
17 September 2018: Professor Vijayakumar headlined the technology section of the Great British Week and delivered a keynote titled "From Automation to Autonomy: The Future of Interactive Robotics". He explored collaborations with Guangdong Province and established links with the new British Consulate outpost led by Mr. David Bull in Shenzhen to support new UK startups trying to expand in China.
10 January 2019: MoU Signed between Turing and RIKEN, The National Institute of Advanced Industrial Science and Technology (AIST) and National Institute of Informatics (NII) Turing News Item

This article was published on 6 Nov, 2024