A behavior-based framework for safe deployment of humanoid robots

Abstract

We present a complete framework for the safe deployment of humanoid robots in environments containing humans. Proceeding from some general guidelines, we propose several safety behaviors, classified in three categories, i.e., override, temporary override, and proactive. Activation and deactivation of these behaviors is triggered by information coming from the robot sensors and is handled by a state machine. The implementation of our safety framework is discussed with respect to a reference control architecture. In particular, it is shown that an MPC-based gait generator is ideal for realizing all behaviors related to locomotion. Simulation and experimental results on the HRP-4 and NAO humanoids, respectively, are presented to confirm the effectiveness of the proposed method.