Object detection and tracking using multi-layer laser for autonomous urban driving

A real-time approach to detecting and tracking multiple objects for an urban driving environment in multi-layer laser data is proposed in this paper. Since situational awareness is crucial for autonomous driving in complicate urban environments, object detection and tracking with cameras or laser has become a popular research topic. With 3D range data, we take the advantage of geometry to cluster point cloud into objects in a fast way. Model-based object tracking framework used in this paper relies on Kalman filter. We set up geometry model for each object and evaluate the observation condition before model updating. We also provide the solution to complicate situations, like splitting, merging and degradation. Our approach has been applied to the multi-layer laser set up on our autonomous driving vehicle under different circumstances, and experiment results are presented.