Pedestrian tracking using Velodyne data — Stochastic optimization for extended object tracking

Abstract

Environment perception is a key enabling technology in autonomous vehicles, and multiple object tracking is an important part of this. High resolution sensors, such as automotive radar and lidar, leads to the so called extended target tracking problem, in which there are multiple detections per tracked object. For computationally feasible multiple extended target tracking, the data association problem must be handled. Previous work has relied on the use of clustering algorithms, together with assignment algorithms, to achieve this. In this paper we present a stochastic optimisation method that directly maximises the desired likelihood function, and solves the problem in a single step, rather than two steps (clustering+assignment). The proposed method is evaluated against previous work in an experiment where Velodyne data is used to track pedestrians, and the results clearly show that the proposed method achieves the best performance, especially in challenging scenarios.