Simulation Video of Socially Aware Mobile Robot Navigation with Human Agents

Le résumé fourni par la source

This video presents a simulation demonstration of a socially aware mobile robot navigating in a human-populated environment. The simulation is conducted in a Gazebo-based environment, where multiple human agents with heterogeneous social behavior profiles interact dynamically with the robot. Human agents are modeled using behavior trees to represent different social attitudes, including curious, neutral, and shy behaviors. These profiles influence how agents respond to the robot’s presence, resulting in diverse interaction patterns such as approach, indifference, and avoidance. The robot operates using a standard navigation stack, while human detection is employed solely as a perceptual trigger to inform behavior-aware navigation, rather than as a learning component. The purpose of this video is to qualitatively illustrate the feasibility of integrating human behavior modeling and socially aware navigation in simulation environments. The demonstration highlights realistic human–robot interaction dynamics without claiming quantitative performance evaluation or real-world deployment. This simulation video serves as supplementary material to support ongoing research on social navigation and human-aware robotic systems.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.