Polar Grid Based Robust Pedestrian Tracking with Indoor Mobile Robot using Multiple Hypothesis Tracking Algorithm

摘要

Autonomous mobile robots manipulating in human environments with the information about where the service target is and in which direction the target is heading typically can improve their interaction behaviors. By providing the capability of tracking the target pedestrian in an appropriate manner, the robot can assist people in various ways under different environments. In this paper, a pedestrian tracking system for indoor single laser range finder mounted robot is presented. The polar occupancy grid based DATMO system critically increases the performance of pedestrian detection. The multiple hypothesis tracking algorithm (MHT) is chose to perform robust pedestrian tracking. The key contribution of paper is the combination of polar grid based DATMO system and MHT which enhances the reliability and robustness of the pedestrian tracking especially when the target moves in complex motion patterns. Several experimental results will be shown to demonstrate the benefits of the proposed idea.