Abstract:
The last half decade has seen a steep rise in the number of contributions on safe learning methods for real-world robotic deployments from both the control and reinforcement learning communities. This talk provides a review of the recent advances made in using machine learning to achieve safe decision-making under uncertainties with a focus on learning-based control approaches that safely improve performance by learning the uncertain dynamics. As data- and learning-based robot control methods continue to gain traction, researchers must understand when and how to best leverage them in real-world scenarios where safety is imperative. This talk highlights some of the open challenges that will drive the field of robot learning in the coming years, and emphasize the need for realistic physics-based benchmarks to facilitate fair comparisons between control and reinforcement learning approaches.
Paper:
"Safe learning in robotics: From learning-based control to safe reinforcement learning," Brunke, Greeff, Hall, Yuan, Zhou, Panerati, and Schoellig, arXiv preprint arXiv:2108.06266, 2021
https://arxiv.org/abs/2108.06266
"safe-control-gym: a Unified Benchmark Suite for Safe Learning-based Control and Reinforcement Learning," Yuan, Z., Hall, A. W., Zhou, Brunke, Greeff, Panerati, and Schoellig, arXiv preprint arXiv:2109.06325, 2021
https://arxiv.org/abs/2109.06325
Safe Robot Learning
https://www.dynsyslab.org/safe-robot-learning/
Bio:
Lukas Brunke is a PhD student at the University of Toronto for Aerospace Studies (UTIAS) and a student affiliate of the Vector Institute, Toronto, Canada. He is a member of the Dynamic Systems Lab where he is supervised by Prof. Angela Schoellig. His research focuses on safe learning-based control with a focus on systems where full-state measurement is unavailable. He received the bachelor’s and master’s degree from Hamburg University of Technology and spent time at the Max Planck Institute for Intelligent Systems, the University of California, Berkeley, and the National University of Singapore.
Find out more at: https://www.lukasbrunke.com/
Toronto AIR Seminar:
The Toronto AI Robotics Seminar Series is a set of events featuring young robotics and AI experts. The talks are given by local as well as global speakers and organized by the Faculty and Students at University of Toronto’s Department of Computer Science. We welcome students, researchers and robotics enthusiasts from around the world to join us and interact with the Toronto Robotics Community.
Find out more at: https://robotics.cs.toronto.edu/toronto-air