Learning Leg Pattern Using Laser Range Finder in Mobile Robots

Abstract

In spite of the advances on people detection and tracking during last years, included the skeleton based trackers, it is interesting to use different types of sensor in this task in order to achieve more robust people detection and tracking algorithms. This work focuses its attention on a laser range finder based approach for people detection and tracking. Patterns of leg are learnt from 2D laser data using machine learning algorithms. Unlike others leg detection approaches, people can be still or moving at the surroundings of the robot. The method of leg detection is used as observation model in a particle filter to track the motion of a person. Then, a Kinect based tracker is proposed to overcome some limitations of laser sensor. Finally both sensors are fused in a multisensor tracker to obtain a robust people detection and tracking system. Experiments on people following in an indoor environment have been used to validate the proposal.