Point-cloud Map Update for Connected and Autonomous Vehicles Based on Vehicle Infrastructure Cooperation: Framework and Field Experiments

Abstract

Constructing and updating of high-definition (HD) map, especially point-cloud map is critical to the deployment and promotion of connected and automated vehicles (CAVs). The prevailing approach is to collect environmental data through mobile data acquisition equipment, and detect changes and update. Existing studies mainly focus on improving the accuracy and efficiency of the mapping algorithm, few of them makes full use of the cooperation between CAVs and road infrastructures. This paper proposes a point-cloud map update approach based on the cooperative vehicle infrastructure system (CVIS), which can reduce the interference on ordinary traffic participants under the premise of ensuring accuracy and efficiency. The main process of the approach is as follows. First, collecting environmental point-cloud data through the LiDAR mounted on the CAV. Then, the CAV sends the point-cloud data to the roadside unit (RSU) through the on-board unit (OBU). The roadside computer unit (RSCU) performs points registration and comparison of cloud data. After that, the RSCU distributes the updated point-cloud map to the CAV within the communication range through the RSU. In addition, corresponding experiments are carried out to verify the feasibility of the proposed approach. Experiment results show that the proposed approach is able to maintain satisfactory performance within a range of at least 500 meters.