Laser-based pedestrian tracking with multiple mobile robots using outdoor SLAM

Abstract

This paper presents laser-based pedestrian tracking with multiple mobile robots in outdoor environments. Each robot finds pedestrians using its own laser scan image by an occupancy-grid-based method, and it tracks the detected pedestrians via Kalman filtering and Global nearest neighbor based data association. In addition, it generates a local map using EKF-SLAM, where cylindrical objects in outdoor environments, such as trees and poles, are used as landmarks. When the robots are located near each other, they exchange their local maps and pedestrians tracking information through intercommunication. Tracking data are combined using Covariance Intersection, and a global map is built by merging the local maps, thus tracking performance can be improved. In our tracking method, all robots share tracking data with each other, so that they can recognize pedestrians that are invisible to any robots. The experimental results of tracking two pedestrians with three mobile robots validate the proposed method.