2020/06/05 by Abhinav Jain, Daphne Chen, Jain, Abhinav +15
Engineering · Psychology · #Autonomous Vehicle Technology and Safety #FOS: Computer and information sciences #Human Motion and Animation #Human-Automation Interaction and Safety #Robotics (cs.RO)
paper · pdf · doi:10.48550/arxiv.2006.03614
openalex publication_date 2020/06/05 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
We address the problem of adapting robot trajectories to improve safety,\ncomfort, and efficiency in human-robot collaborative tasks. To this end, we\npropose CoMOTO, a trajectory optimization framework that utilizes stochastic\nmotion prediction models to anticipate the human's motion and adapt the robot's\njoint trajectory accordingly. We design a multi-objective cost function that\nsimultaneously optimizes for i) separation distance, ii) visibility of the\nend-effector, iii) legibility, iv) efficiency, and v) smoothness. We evaluate\nCoMOTO against three existing methods for robot trajectory generation when in\nclose proximity to humans. Our experimental results indicate that our approach\nconsistently outperforms existing methods over a combined set of safety,\ncomfort, and efficiency metrics.\n