Abstract: The sensor model is an important part of intelligent vehicle virtual verification. In this paper, the real-time functional models of millimeter wave radar, lidar and camera are established for the requirements of intelligent vehicle verification in large-scale virtual simulation environment. Firstly, the model develops an effective management and fast search engine for simulation data based on the gridding method, and studies a fast object extraction method to meet multi-vehicle sensor requirements in large-scale simulation environment. Secondly, a fast algorithm for judging the occlusion relationship between objects in a two-dimensional plane is studied based on the shadow volume algorithm, and the occlusion effect between objects is simulated. Thirdly, according to the outputs of the sensors, the object coordinates in the inertial coordinate system are converted to the sensor coordinates to form a standard output. Finally, a noise simulation model is established according to the noise characteristics of sensors. The models of Delphi ESR millimeter wave radar, Ibeo four-line lidar and Delphi IFV300 monocular camera are established in the virtual test platform. The function and performance of the models are simulated and verified. The results show that the sensor functional models established in this paper can accurately simulate the detection of obstacles, and the detection accuracy meets the requirements. The simulation time achieves sub-millisecond on normal computers and the simulation meets real-time requirements

Key words: intelligent vehicle, virtual test, sensor, functional model, verification