Font Size: a A A

Research On Control Strategy Of Active Collision Avoidance Of Intelligent Vehicle

Posted on:2021-04-23Degree:MasterType:Thesis
Country:ChinaCandidate:Y J SunFull Text:PDF
GTID:2392330629987109Subject:Vehicle engineering
Abstract/Summary:PDF Full Text Request
As an indispensable part of vehicle active safety,the active collision avoidance system has been widely concerned by automobile enterprises and research institutes in recent years,which can effectively reduce collision accidents.However,most of the current active collision avoidance system can not make comprehensive use of longitudinal braking collision avoidance and steering collision avoidance for reasonable collision avoidance planning,which is difficult to deal with the complex driving conditions of the intelligent vehicle.Therefore,by introducing the multi-agent theory,the multi-agent active collision avoidance system for intelligent vehicle was established.By combining longitudinal collision avoidance and steering collision avoidance,the intelligent vehicle can choose reasonable collision avoidance methods under different working conditions.And according to the predictive control algorithm,the intelligent vehicle following collision avoidance controller and the emergency steering collision avoidance path following controller were designed,which effectively improved the active collision avoidance ability of the intelligent vehicle.The main research contents are as follows:Firstly,the research on active collision avoidance at home and abroad was analyzed,and the multi-agent system and multi-agent conflict problem were introduced.Aiming at the shortcomings of the current active collision avoidance system,the multiagent active collision avoidance control system of intelligent vehicle multi-agent was proposed,and the research ideas and chapter arrangement were also introduced.Then,the multi-agent active collision avoidance system for intelligent vehicle was established,in which three sub-agents were established,namely,the vehicle following collision avoidance agent,the emergency braking collision avoidance agent and the emergency steering collision avoidance agent.The model of the fixed time headway from vehicle following was established in the vehicle following collision avoidance agent;In the emergency braking collision avoidance agent,the braking safety distance model under three emergency conditions was established,which took into account the road adhesion and the time delay of each link.In the emergency steering collision avoidance agent,by considering the road adhesion characteristics and vehicle dynamics characteristics,the path of the emergency steering collision avoidance was determined based on quintic polynomial.At the same time,the blackboard model was introduced to coordinate and control the decision conflicts generated by the above agents in collision avoidance,so that the intelligent vehicle can choose a reasonable collision avoidance method under different working conditions.And then,according to the predictive control theory,the following collision avoidance controller and the emergency steering collision avoidance path following controller were designed.Emergency steering avoidance had a great advantage under the driving conditions of high speed and low adhesion road surface.Therefore,the soft restraint control of tire side deflection angle was added in the path following controller to ensure that the tire side deflection angle and side deflection force was always kept in a linear range,which avoided the large sideslip phenomenon in the course of lane changing and collision avoidance.The effectiveness of the emergency collision avoidance strategy and the predictive controller was verified by building the collision avoidance scenario in CarSim and the control model in Simulink for joint simulation.Finally,in order to verified the effectiveness of the emergency braking safety distance model,a real vehicle collision avoidance experiment was carried out.Firstly,the data obtained by the intelligent vehicle through millimeter wave radar and camera were fused,and the obstacle screening algorithm was designed.Then the emergency braking collision avoidance controller and actuator were selected,and the braking collision avoidance control program was written.Finally,the effectiveness of the safe distance model for emergency collision avoidance was verified by vehicle experiments,which will provide reference for further research on the active collision avoidance system.
Keywords/Search Tags:active collision avoidance system, multi-agent theory, safe distance model, conflict resolution, predictive control
PDF Full Text Request
Related items