Dynamical System based Socially-aware Navigation for Mobile Robot

Abstract

This paper presents a novel framework for socially-aware robot navigation in a pedestrian mixed environment. To improve the acceptance of mobile robots in daily life, robots should exhibit safe and socially acceptable motions to humans, as integrating social norms has been shown to contribute to comfort and socially-aware behavior in various studies. In this paper, we propose a novel framework, Socially-aware Dynamical System based Control (Social-DSC), which leverages the Dynamical System based Control (DSC) for collision avoidance and Social Force Model (SFM) for socially-aware robot navigation. While safety is guaranteed by the DSC framework, the social norms are considered in the modified SFM. In simulation studies, we compared the proposed method with baseline methods, demonstrating higher pedestrian path smoothness and reduced effort in evading motion, thereby improving human comfort.