Research On Vehicle Active Collision Avoidance Control Based On Braking And Steering

With the continuous growth of automobile ownership,traffic accidents are also increasing.In order to reduce traffic accidents and improve driving safety,active collision avoidance control of automobiles has gradually become a hotspot research in various countries in recent years.This paper studies the active collision avoidance control by braking and steering.According to the different collision avoidance methods,the research is mainly carried out from the four aspects which are braking and steering safety distance model,braking and steering switching strategy,path planning,and trajectory tracking based on model predictive control.Firstly,the overall design of vehicle active collision avoidance control is carried out,according to the AEB test condition of E-NCAP,a simplified straight road test scenario is designed,and the T-junction merging test scenario is added according to the accident statistics data.Then by analyzing the longitudinal distance required for braking and steering,the critical speed for switching between braking and steering is obtained.Based on the critical speed,the switching strategy for braking and steering collision avoidance is established with TTC as the trigger condition.In the aspect of active braking control,four safety distance models,TTC,Honda,Mazda and Berkeley,are simulated and analyzed.The TTC and Honda models,which are involved later and have smaller stopping distance,are selected as the safe distance models of braking control.At the same time,the parameters of TTC model are optimized,and the safety distance model which covers 0-80km/h is obtained.In terms of path planning,a comparative analysis of common car path planning algorithms was carried out,the five order polynomial planning with simple expression and fast calculation speed is selected as the path planning method for active steering avoidance.Then,a simplified scenario of lane change with two lane is established,and the safety distance model for lane change is derived according to the simplified scenario.In the aspect of path tracking,a linear model predictive controller is designed,which takes the model predictive control as the path tracking algorithm of lane changing,and it brings the vehicle dynamics constraints into the quadratic programming solution,so that the controller can not only meet the control accuracy,but also achieve the control smoothness.In this paper,a complete active collision avoidance joint simulation software platform is built based on Prescan,Carsim and Matlab/Simulink.The critical safety distance between braking and steering is analyzed,and the vehicle's active collision avoidance control strategy based on vehicle speed and TTC is obtained.The established collision avoidance control strategy is tested in different test conditions.The test results show that the car can avoid collision well at different speed and low friction coefficient road.