Adaptive collision sensitivity determination based on link effective mass

Lukas Rustler, Matej Misar, and Matej Hoffmann propose an adaptive collision sensitivity framework: collision forces are dynamically estimated according to the effective mass (EM) of each robot link, and whether to interrupt motion is judged on that basis, instead of using a fixed contact threshold for the whole robot. The authors conducted collision detection and isolation experiments on a real UR10e equipped with AIRSKIN sensitive skin and in simulation, and supplemented them with simulation experiments on a KUKA LBR iiwa based on joint torque sensing. The authors report that in simulated pick-and-place scenarios containing transient and quasi-static collisions, production efficiency improved by more than 45% compared with the standard method; the method is applicable to any robot for which effective mass can be computed. The above improvement comes from the authors' reports in simulated scenarios and has not yet been validated in real production environments.

AI 根据报道生成 · 2 days ago更新