SRPB: a benchmark for the quantitative evaluation of a social robot navigation

摘要

In recent years, social robots have gained popularity and many human-aware navigation approaches have emerged. This work presents SRPB – a comprehensive benchmark for the quantitative evaluation of robot navigation methods. It is intended to assist the system designer in selecting the best method for the application. We tested traditional and human-aware trajectory planners with the TIAGo robot in simulated and real-world environments (Fig. 1). We assessed robot behaviour using state-of-the-art task performance scores and novel social metrics regarding robot motion naturalness and the perceived safety of humans surrounding the robot. Our social metrics take human tracking reliability into account. An open-source implementation of our benchmark, compatible with the Robot Operating System, is provided.