For autonomous driving, accurate detection and recognition of the surrounding environment are essential to ensure the safety driving, and then Light Detection And Ranging has been used as an external recognition sensor, which plays a role in obstacle recognition. However, the problem with LiDAR is that the point cloud data is sparse. This paper proposes a method to obtain the depth information of feature points from objects motion by using the optical flow method and fuse them with the LiDAR point cloud data in order to improve the detection capability.

View full abstract