作者: Matteo Saveriano , Fabian Hirt , Dongheui Lee
DOI: 10.1016/J.PATREC.2017.04.014
关键词:
摘要: In this work, we present a real-time approach for human-aware motion replanning using two-level hierarchical architecture. The lower level leverages stable dynamical systems to generate motor commands and online reshape the robot trajectories. reshaping strategy modifies velocity of match three requirements: i) increase human safety in case close interaction with robot, ii) guarantee correct task execution unforeseen obstacles (including human), iii) replan current taking into account behavior. has execute all needed computations real-time. To end, also propose novel that depth space structure parallel programming rapidly accurately estimate robot-obstacle distances. higher architecture monitors provides information about status. proposed been tested scenario, showing promising results terms safe human-robot coexistence.