Research On Avoiding Collision Algorithm Based On V2P

The rapid increase in car ownership leads to frequent traffic accidents,in which the accidents between vehicles and pedestrians are particularly serious.Domestic and foreign scholars and experts have done a lot of research work on collision avoidance between vehicles and pedestrians,but they have not considered different control capabilities for the drivers and different characters for the pedestrians.Therefore,the thesis studies the driver’s behavior and evaluates the driver’s comprehensive quality based on safety.Then a traffic intersection is selected as a specific scene,and the collision avoidance algorithm is studied for different types of drivers and pedestrians based on game theory.The contents of work are as follows:1.Research on the method of evaluate driver’s comprehensive quality based on CAN bus data.In the research,a real-time vehicle CAN data collection platform is developed,and a model for evaluating the driver’s comprehensive quality is established based on traffic safety,ride comfort and vehicle fuel consumption.Firstly,the evaluation parameters are extracted from the CAN bus data to construct the evaluation system.To obtain the weight of each evaluate parameter,the fuzzy analytic hierarchy process is improved by using the entropy method combined with the vector norm.Secondly,Using the fuzzy theory,the membership function of each evaluate parameter is established,and fuzzy evaluation matrix is obtained.In the end,according to the principle of maximum membership degree,the driver’s comprehensive quality is graded and evaluated.Actual test data proves that the model can accurately evaluate the comprehensive quality of the driver and alert the driver to improve driving behavior.2.Research on collision avoidance algorithm based on V2 P at intersection.In the research,no signal intersection is selected as a specific scene,and a V2 P collision avoidance algorithm based on game theory is studied to give optimal traffic strategies for different types of vehicles and pedestrians to avoid collision.Firstly,delay payment and collision risk payment of vehicles and pedestrians are quantified at intersection,and a traffic game model for vehicles and pedestrians is established.Secondly,the vehicle rewards and penalty payments are introduced to improve the game model.According to set up simulation parameters of the experimental environment,the results of the game model before and after the improvement are contrasted and analyzed.It is proved that traffic rewards and penalties are introduced to increase the vehicle’s rate of passing to the pedestrians.Finally,the idea of virus propagation model is used to analyze traffic participants’ violation.The numerical simulation further proves the necessity of introducing traffic reward and punishment to the avoidance algorithm.This algorithm provides a theoretical reference for the development of collision avoidance systems for vehicles and pedestrians in the future.