Autonomous Navigation System Test for Service Robots in Semi-Structured Environments

This paper presents an autonomous mobile service robot designed by the Conscious Technologies research group at Tecnologico de Monterrey, integrating advanced navigation and real-time surveillance capabilities. The primary objective was to evaluate the robot’s performance in both navigation and vision tasks within dynamic environments. The research compared two navigation algorithms: the Navstack package, and a State Feedback Controller (SFC). While the Navstack algorithm excelled in obstacle avoidance, it showed variability in location precision and trajectory repeatability. Conversely, the SFC demonstrated superior precision and repeatability, but lacked obstacle-detection capabilities. Furthermore, the vision system’s efficacy in face mask detection and social distancing compliance was assessed under varying robot and human speeds. An ANOVA analysis revealed the significant impact of these dynamic variables on the performance of the vision model. The integration of robust navigation algorithms with intelligent surveillance methods highlights the potential for real-world applications in protocol enforcement and autonomous navigation in semi-structured environments.