SELFI: Autonomous Self-Improvement with RL for Vision-Based Navigation around People

Noriaki Hirose, Dhruv Shah, Kyle Stachowicz, Ajay Sridhar, Sergey Levine

Keywords: online reinforcement learning, vision-based navigation

TL;DR: Online model-free RL coorperating offline model-based learning objective and its application to vision-based navigation

Abstract: Autonomous self-improving robots that interact and improve with experience are key to the real-world deployment of robotic systems. In this paper, we propose an online learning method, SELFI, that leverages online robot experience to rapidly fine-tune pre-trained control policies efficiently. SELFI applies online model-free reinforcement learning on top of offline model-based learning to bring out the best parts of both learning paradigms. Specifically, SELFI stabilizes the online learning process by incorporating the same model-based learning objective from offline pre-training into the Q-values learned with online model-free reinforcement learning. We evaluate SELFI in multiple real-world environments and report improvements in terms of collision avoidance, as well as more socially compliant behavior, measured by a human user study. SELFI enables us to quickly learn useful robotic behaviors with less human interventions such as pre-emptive behavior for the pedestrians, collision avoidance for small and transparent objects, and avoiding travel on uneven floor surfaces. We provide supplementary videos to demonstrate the performance of our fine-tuned policy.

Video: https://www.youtube.com/watch?v=ElTAc_9a2l4

Website: https://sites.google.com/view/selfi-rl/

Student Paper: no

Submission Number: 338