Environmental Sensors For Intelligent Driver Assistant System Research On Evaluation Methodology

With the continuous development of intelligent assistant driving system technology,people pay more and more attention to its technological,convenience and safety.Intelligent assistant driving system mainly includes adaptive cruise control(ACC),autonomous emergency braking(AEB),forward collision warning(FCW),lane keeping assistant function(LKA),lane departure warning(LDW)and other functions.Sensors are an important part of all assistant driving systems.If there is no good sensor to provide reliable environmental information,all assistant driving systems will become unreliable.Sensor schemes used in intelligent assisted driving generally include single camera,single radar,or the combination of the two,but they all output the recommended target-level information.This paper evaluates the impact of the accuracy of target-level information on the assisted driving system.Therefore,this paper is mainly divided into three parts: The first part is to establish the mainstream dynamic control model of assistant driving function,and establish corresponding control models for different assistant driving functions(including ACC,AEB,LKA).The second part is the analysis of the sensor information module used in each assistant driving function,such as dangerous target screening module,safe distance calculation module,speed or distance control switching module,AEB including dangerous target screening module,AEB function triggering module,LKA turning angle decision module for different curvature radius and so on.The information output of sensors is different,and the evaluation conditions are formulated,and reasonable evaluation indicators are formulated.The third part is the evaluation method of sensors.This part mainly aims at the different information used by sensors in different assistant driving functions.Analytic Hierarchy Process(AHP)is used to analyze the weight of each sub-module of assistant driving function or evaluation condition in the sensor evaluation.Then the weight of different working conditions in each sub-module is analyzed.The weights in the rating conditions are analyzed,and a complete set of sensor evaluation criteria is finally formed.In order to validate the proposed evaluation method,this paper combines simulation with real vehicle.Firstly,a CarSim and MATLAB/Simulink joint simulation test platform is built to validate the various assistant driving functions required for sensor testing.Then,a real vehicle platform is built,and a camera and a millimeter wave radar are installed on the existing test vehicle to transmit the target level after the integration of the two.The output data is recorded by the sensor data recording and visualization software developed by the research group,and then the experiment is carried out.The measurement accuracy of the sensor is analyzed by measuring the static targets at different locations in front.Then the sensor error is equivalent to the simulation model,and the sensor is evaluated by the final simulation result.