The study of traffic flow model based on cellular automata and Naive Bayes

Abstract

A new traffic flow model is proposed based on cellular automata and Naive Bayes theory to effectively describe the traffic flow velocity and flow state of the road. On the basis of NaSch model, the safety distance is fully considered in this model, and random deceleration and inflow rules of a vehicle are introduced. At the same time, vehicle acceleration, deceleration and lane change are optimized with Naive Bayes theory. Finally, experimental platform is used for numerical analysis, and the relationship between such parameters as average velocity of traffic flow, maximum velocity of traffic flow, number of inflow vehicles and random deceleration probability, etc. is studied in depth. The results show that the maximum velocity has a great effect on the traffic flow state, and when the vehicle inflow probability is lower, random deceleration probability has less effect on the average velocity and the number of vehicles waiting for inflow; on the contrary, the higher the random deceleration probability is, the more obvious the tendency of road congestion.