Moving-object Tracking with Lidar Mounted on Two-wheeled Vehicle

This paper presents a tracking (estimating position, velocity and size) of moving objects, such as cars, two-wheeled vehicles, and pedestrians, using a multilayer lidar mounted on a two-wheeled vehicle. The vehicle obtains its own pose (position and attitude angle) by on-board global navigation satellite system/inertial navigation system (GNSS/INS) unit and corrects the distortion in the lidar-scan data by interpolating the pose information. The corrected lidar-scan data is mapped onto 3D voxel map represented in the world coordinate frame. Subsequently, the vehicle extracts the interested lidar-scan data from the current lidar-scan data using the normal distributions transform (NDT) scan matching based map-subtraction method. The extracted scan data are mapped onto an elevation map, and moving objects are detected based on an occupancy grid method. Finally, detected moving objects are tracked based on the Bayesian Filter. Experimental results show the performance of the proposed method.