Evaluating Reinforcement Learning for Human-Robot Handovers Ben Gurion University of the Negev, Israel October 2019 - December 2022 Collaborators: Alap Kshirsagar (Cornell), Guy Hoffman (Cornell), Tair Faibish (BGU), Armin Biess (BGU)

This project studies model-based reinforcement learning (RL) to train a robot controller for human-robot object handovers. RL is a promising approach to develop handover policies, but existing methods did not consider important aspects of human-robot handovers, namely large spatial variations in reach locations, moving targets, and generalizing over mass changes induced by the object being handed over. We report on promising benefits, but also limitations of existing RL methods for this HRI task.

Publications

*Kshirsagar, A., *Faibish, T., Hoffman, G. & Biess, A., Lessons Learned from Utilizing Guided Policy Search for Human-Robot Handovers with a Collaborative Robot, International Conference on Robotics, Automation and Artificial Intelligence (RAAI), 2022

Kshirsagar, A., *Hoffman, G., & *Biess, A., Evaluating Guided Policy Search for Human-Robot Handovers, IEEE Robotics and Automation Letters, and IEEE International Conference on Robotics and Automation (ICRA), 2021