Research On Collaborative Decision Model Of Highway Emergency Obstacle Avoidance In Intelligent And Connected Environment

With the continuous development of the society and economy in the past 40 years of reform and opening up,transportation volume of passengers and freight by road has dramatically increased.As the dominant way to undertake long-distance road transportation,highways are facing the problem of excessive traffic volume.Meanwhile,the safety level under large traffic is also facing increasing risks.Intelligent transportation system(ITS)is brought out as an effective solution for these problems,and its ultimate objective is to achieve an integrated system of vehicles and infrastructures,and to realize automatic driving within the system.This proposal studied the highway emergency obstacle avoidance scene in intelligent and connected environment,and proposed a decision model for multi-vehicle collaborative control using the artificial potential field(APF),to ensure safe and efficient operation of highways under emergency obstacle avoidance scenarios.Firstly,this proposal analyzes the composition of the integrated system of vehicles and infrastructures and elaborates functions of components of the system based on existing studies.The system includes cloud subsystem,roadside subsystem,and on-vehicle subsystem.Meanwhile,the hypothesis premise of this paper is proposed for the research scene,and the process of perception,decision making and control of the system is expounded.Secondly,this proposal takes driving characteristics of highway vehicles into consideration,i.e.the willingness to maintain the desired speed,keep along the lane center,avoid approaching road boundary and avoid approaching other obstacles,including other vehicles,and the APF strength model is established in the intelligent and connected environment.The model includes four kinds of potential fields: speed potential,lane potential,road boundary potential,and physical obstacle potential.Functional expressions for each kind of potential are proposed respectively.Thirdly,the longitudinal and lateral control decision model of a single vehicle under APF is established according to the distribution and variation of APF near the vehicle position.Longitudinal accelerations and lateral lane change decisions can be made by the cloud according to the APF model to achieve autonomous driving.In addition,a multi-vehicle collaborative decision model is proposed based on the single-vehicle control model,to take full advantage of the intelligent and connected environment and coordinately control multiple vehicles to avoid obstacles safely and efficiently.Then,a microscopic simulation environment is built using Python for numerical simulation.By comparing the highway emergency obstacle avoidance multi-vehicle collaborative decision model and the lane change model under artificial driving environment,reasonable indicators are selected to analyze the performance in terms of efficiency,safety,and comfort.The experimental results show that the multi-vehicle collaborative decision model proposed in this proposal has obvious superiority to the artificial driving environment,and can significantly improve the capacity,flow speed,safety and comfort under emergency obstacle avoidance scenes.Finally,the research content of this paper is summarized,the insufficiency of the research are analyzed,and the future research direction is prospected.