People tracking with in-vehicle multi-laser range sensors

Abstract

This paper presents a method for people tracking with in-vehicle 2D laser range sensor (LRS). Since a view angle of the LRS is limited in orientation, mobile robot is equipped with multi-LRS's for omnidirectional sensing. In order for people tracking by multi-LRS's cooperation, the coordinate frames of multi-LRS's are calibrated based on Kalman filter and chi-hypothesis testing. People tracking is based on Kalman filter, Covariance Intersection and the assignment algorithm. A rule based track management system is embedded into the tracking system in order to improve the tracking performance. The experimental result for three people tracking in indoor environments validates the proposed method.