Compliant and Versatile Robot Control in Human Environments:
|
Despite tremendous advances in robotics, we are still far behind having robotic systems that can safely, adaptively, and reliably operate in human environments. Learning, adaptation, reaction and compliant control are vital elements to enhance robot manipulation performance and safety in unstructured environments. It enables robots to extend their skills beyond a set of hardcoded trajectories, while allowing safe humanrobot coexistence and better handling of perturbations and interactions with the environment. Currently, robot learning and compliant control are mainly studied in the two largely disconnected domains of machine learning and control theory, respectively. This workshop aims at bridging the gap between these two research fields in order to achieve compliant and versatile robots that can truly function in human environments. Robots with such capabilities can be extremely useful in various applications such as rehabilitation, assistive robots, human-robot collaboration, teleoperation as well as industrial and home service robotics. In this half a day workshop, experts in the field of machine learning and robot control will present their latest achievements and discuss possible directions to integrate the various techniques required to create versatile robots. |
Important Dates
|
Due to its interdisciplinary nature, this workshop is supported by two IEEE RAS Technical Committees: Human-Robot Interaction & Coordination and Robot Learning. The workshop's topics cover, but are not limited to: