Study On Optimization Of Adaptive Cruise Control Strategy For Cut-in Scenario

ADAS(Advanced Driver Assistance System)is one of the important directions of the current automotive industry,as the basic function of the ADAS,the system architecture,control algorithm and related products of adaptive cruise control(ACC)are basically mature.However,in the case of the cut-in scenario,there usually exists late brake,acute brake or brake intervention which bring people poor driving experience.Therefore,this paper is proposing an optimization research of adaptive cruise control strategy for cut-in scenario.First,design and develop an adaptive cruise control algorithm model.The control algorithm model is divided into the perception layer,the decision layer and the execution layer.The perception layer establishes the perception model based on millimeter wave radar;the control algorithm model of the decision layer is designed based on the cruise mode and the following mode which constructs the performance function based on MPC theory;the execution layer establishes an lower model based on vehicle longitudinal dynamics.Last,build the ACC control algorithm model based on Simulink & Carsim software platform to verify the effectiveness of the algorithm in typical ACC test cases.Secondly,in the case of excessive braking intervention of the ACC system in the cut-in scenario,the personalized cut-in target switching strategy based on driver's characteristics is carried out.First,the driver's driving characteristics test is carried out.According to the time to collision and following time gap during the test,the driving characteristics of drivers are analyzed,then a driver type classifier is designed based on the support vector machine which provides a hyperplane using SMO algorithm with relaxation variable and penalty factor to identify the type of driver.Secondly,the cut-in target switching strategy according to the distance and relative velocity is designed using least squares to optimize the ACC control target update mechanism for different types of drivers.Last,verify the effectiveness of optimized strategy with Simulink and Carsim joint simulation.Thirdly,in the case of late braking and acute braking in the cut-in scenario,An optimization algorithm combining potential target selection and cut-in vehicle pre-identification is proposed,as well as a strategy of deceleration optimization based on analysis of collision threshold to improve the driving comfort from the driving safety.First,the potential target selection expands the selecting vision to monitor and track the state of targets from adjacent lane.Then,cut-in vehicle pre-identification algorithm quantifies the probability of cutting-in behavior.Next,the optimization of deceleration for cut-in scenario based on analysis of collision threshold is proposed.Last,based on the Simulink-Carsim co-simulation,verify the effectiveness of the optimization.Last,a real vehicle test platform is built based on Micro Autobox,first establish interfaces of controller and hardware,then calibrate the automatic throttle and brake of the vehicle Last,based on the real vehicle test platform,the test is carried out in the main urban area of the city to verify the effect of the adaptive cruise control algorithm for cut-in scenario which is proposed in this paper.