Optimal Configuration Scheme of Multi-Sensor Perception for Autonomous Vehicles based on Solid-State LiDAR

Environmental perception is the first and most critical part of realizing autonomous driving. Solid-state LiDAR, as an important sensor in the perception layer of autonomous vehicles, gives the vehicles the ability to perceive beyond human eyes and guards the safety of people's travel. This paper proposes an optimal multi-sensor configuration method for autonomous vehicles based on solid-state LiDAR, camera, millimeter wave radar, and ultrasonic radar. The range of interests (RoIs) is divided into six zones and they are further classified into the core RoI and the extended RoI based on importance. The RoIs discretization mechanism is designed according to the functions and sensing range of various sensors. Considering the cost, coverage, and redundancy, a mixed integer programming model is built to determine the optimal number of sensors and placement positions. The optimal configuration scheme is obtained by using the IBM ILOG CPLEX solver. A model of a well-known vehicle brand is chosen to verify the effectiveness and feasibility of the proposed method. The solid-state LiDAR-based multi-sensor perception solution can meet the needs of environment perception for autonomous vehicles, and its coverage and redundancy are better than other control schemes, but its cost is relatively high due to the high cost of the solid-state LiDAR. The configuration scheme proposed in this paper will provide effective support for the design of an autonomous vehicle sensing solution.