Detection and Tracking of Moving Objects Using 2.5D Motion Grids

Autonomous vehicles require a reliable perception of their environment to operate in real-world conditions. Awareness of moving objects is one of the key components for the perception of the environment. This paper proposes a method for detection and tracking of moving objects (DATMO) in dynamic environments surrounding a moving road vehicle equipped with a Velodyne laser scanner and GPS/IMU localization system. First, at every time step, a local 2.5D grid is built using the last sets of sensor measurements. Along time, the generated grids combined with localization data are integrated into an environment model called local 2.5D map. In every frame, a 2.5D grid is compared with an updated 2.5D map to compute a 2.5D motion grid. A mechanism based on spatial properties is presented to suppress false detections that are due to small localization errors. Next, the 2.5D motion grid is post-processed to provide an object level representation of the scene. The detected moving objects are tracked over time by applying data association and Kalman filtering. The experiments conducted on different sequences from KITTI dataset showed promising results, demonstrating the applicability of the proposed method.