Title: Compliant and Stable Robot Control for Physical Human-Robot Cooperation
Abstract: Robots have widespread use in manufacturing where they operate in highly structured environments with minimal and heavily controlled human-robot interaction. They have also been introduced to other industries (e.g.,\ health care), which can require more dynamic and interactive tasks that cannot be fully characterised a priori. The impedance controller is a widespread technique enabling robots to interact with uncertain environments within certain boundaries, which defines the relationship between the generated force by the robot and its location in its surrounding environment. This control technique relies on the inverse dynamics modelling to drive the robot to act with an equivalent mechanical impedance or as a Mass-Spring-Damper system, nevertheless, the stability of such systems highly depends on pre-tuned controller gains, which are difficult to obtain for dynamic tasks (e.g.,\ polishing, locomotion, etc.). This is more evident in unstructured environments that require adaptive trajectories and/or variable impedance gains. A particular set of tasks to be highlighted are those where uncertain end-effector contact, against other agents or the environment, may occur (e.g.,\ polishing, human-robot cooperation, etc.) both in proximate and remote human-robot cooperation/collaboration/interaction scenarios. The proposed controller can deal with uncertainties in various human/environment-robot interaction scenarios, using the passivity framework to guarantee safety and stability without relying on additional virtual energy tanks. With this proposed controller, there will be no need of re-tuning the controller gains by the human operator before/after every arbitrary task execution and this can simply be done by the robot itself, by setting its controller gains to be adaptable.