Ring attractor bio-inspired neural network for robot social navigation

We introduce a bio-inspired navigation system for a robot to guide a social agent to a target location while avoiding static and dynamic obstacles. Our system integrates sensory representation, decision-making, and motor control with a novel architecture that uses ring attractor networks. The system controls the robot's direction and speed to keep the follower within a certain distance range. Our approach is compared to the widely-used Social Force Model using the Social Individual Index and Relative Motion Index as metrics in simulated experiments. The results show improved performance with our proposal compared to the Social Force Model.