Development Of Self-driving Virtual Test Platform Based On Human-vehicle-environment Modeling

In the automotive industry,the development of self-driving vehicles is becoming more and more eye-catching.However,autonomous driving requires a large number of road test mileage.At present,there are few automatic driving test sections developed at home and abroad,and the test scene is limited,which is difficult to meet the test requirements of self-driving vehicles.The self-driving virtual test platform restores the real vehicle test conditions maximumlly through vehicle simulation,scene simulation,sensor simulation,driving simulator and controller integration.Therefore,the self-driving virtual test platform can partially replace the road test,complementing the road test,and at the same time,it can greatly shorten the test cycle,reduce the test cost,and protect the life safety of testers.In this paper,a self-driving virtual test platform based on human-vehicle-environment modeling is built,and the feasibility of the platform is tested by human-machine integrated control strategy.The research content mainly includes the following aspects: Firstly,the virtual test scene is built by using the visual driving software Vega Prime and the 3D modeling software 3Dmax/Creator.The software system has the advantages of high real-time performance and high image realism.Then,a real-time system module based on NI-PXI is built.The function of this module is mainly realizing the functions of vehicle simulation and sensor simulation.In addition,the program design of the controller module is built,the human-machine co-drive switching module is built,and the Logitech G27 driving simulation device is integrated into the controller system by using Motohawk.Furthermore,the integration and debugging of the entire virtual test platform mainly includes integrated debugging of hardware and software.The hardware integrated debugging is mainly the signal acquisition of the Logitech G27 driving simulation equipment.Software integration debugging is mainly the design of communication systems.Finally,the test and analysis of the automatic driving virtual test platform is realized.This paper takes the human-machine integrated driving control strategy as an example to test the feasibility of the platform.The test conditions in which the automatic driving system cannot handle the driving environment and request the driver to operate the vehicle are designed,analyzing the driver’s operating behavior.Through experimental analysis,the self-driving virtual test platform built in this paper can fully test the control strategy of self-driving cars.