Research On Evaluation Method Of Shared Human-machine Co-driving System

Automobile intelligence is gradually becoming the development trend of the automobile industry.However,due to technology and regulations,there is still a big gap between the realization of fully automated driving,and human-machine co-driving is still the main form of current intelligent vehicles.Test and evaluation is an important part of the development of human-machine co-driving system.Although domestic and foreign scholars have made a lot of research on the test and evaluation methods of human-machine co-driving system,there are no system level and vehicle level test and evaluation methods at present..Based on the research results of human-machine codriving system at home and abroad,this paper explores and studies the evaluation method of shared human-machine co-driving system.The main research contents are as follows:Firstly,a lateral shared human-machine co-driving system is built,including:arbitrary road following control driver model,model prediction controller and humanmachine weight distribution fuzzy controller to realize the lateral control of vehicles;The longitudinal shared human-machine co-driving system is built,including expected acceleration calculation model,longitudinal driver model,auxiliary controller model and human-machine weight distribution model to realize the longitudinal control of the vehicle.Secondly,the function of shared human-machine co-driving system is analyzed.Combined with the existing intelligent vehicle test and evaluation methods,the test conditions of lateral and longitudinal control ability of human-machine co-driving system are established.According to the characteristics of shared human-machine codriving system,considering the active safety and driving comfort of intelligent vehicle,the evaluation index of shared human-machine co-driving system is proposed,and the evaluation index system of shared human-machine co-driving system is established.Then,the analytic hierarchy process is used to determine the relative weight of each evaluation index in the evaluation index system.Based on CarSim/Simulink,a typical working condition for testing vehicle lateral and longitudinal control performance is established.The built human-machine co-driving system is simulated and tested,and the test results are analyzed and processed,The specific values of the evaluation indexes of the human-machine co-driving system are obtained.Finally,according to the reasonable value range of each evaluation index,the scoring standard of each evaluation index is established,the fuzzy comprehensive evaluation model of shared human-machine co-driving system is constructed by using the fuzzy comprehensive evaluation method,the evaluation of the built shared humanmachine co-driving system is completed,and the evaluation method of shared humanmachine co-driving system is put forward.A shared human-machine co-driving system model is built in Prescan/Simulink to verify the proposed evaluation method of shared human-machine co-driving system.The results show that the proposed evaluation method is suitable for different shared human-machine co-driving systems,and the evaluation results are reliable.Figure[32]Table[20]Reference[80]...