[ICRA25] Proactive Assistance in Human-Robot Collaboration through Task Progress Estimation скачать в хорошем качестве

[ICRA25] Proactive Assistance in Human-Robot Collaboration through Task Progress Estimation

This video summarizes our recent work on proactive assistance for human-robot collaboration. Our Proactive Assistance through action-Completion Estimation (PACE) framework incorporates a novel method that combines Dynamic Time Warping (DTW) with correlation analysis to track human task progression from hand movements. PACE trains a reinforcement learning policy from limited demonstrations to generate a proactive assistance policy that synchronizes robotic actions with human activities, minimizing idle time and enhancing collaboration efficiency. This video also shows the real-world experiments where a manipulator proactively assists a human to assemble an IKEA chair. This work was done by Davide De Lazzari (UniPD), Matteo Terreran (UniPD), Giulio Giacomuzzo (UniPD), Siddarth Jain (MERL), Pietro Falco (UniPD), Ruggero Carli (UniPD), Diego Romeres2 (MERL).