Laser-Based Pedestrian Tracking by Multi-Mobile Robots

Abstract

This paper presents a laser based pedestrian tracking by multi-mobile robots in vicinity. Each robot finds moving people (pedestrians) from its own laser scan image based on an occupancy-grid-based method, and it tracks the detected pedestrians via Kalman filtering and Global-nearest-neighbor-based data association. The tracking data are broadcasted to multi-robots through intercommunication, and they are blended based on Covariance Intersection. By our tracking method, all the robots share the data of tracked pedestrians with each other, so that they can always recognize the pedestrians that are invisible to any robots. The simulation and experimental results validate the proposed method.