Research On Road Confluence Control Method Of Intelligent Vehicle Based On Model Prediction

Intelligent vehicle utilizes the technologies of environmental perception,behavioral decision-making,intelligent control et.al.to automatically conduct common driving operations and reduce traffic accidents which are caused by driver fatigue and misjudgment.Lane changing and lane keeping are two of the most common operations during vehicle driving process.Lane changing is also one of the most likely operations to cause traffic accidents and road congestion.Therefore,it is of certain urgency and practical significance to study the intelligent vehicle lane changing technology.In this paper,the intelligent vehicle is studied for the lane merge(road confluence)scene,aims to design a parallel control algorithm based on model predictive control.The research content of this paper mainly includes four aspects:The merge strategy and the trajectory are proposed.This paper calculate the latest lane change position according to the minimum driving safety distance formula,and selects the timing of the lane change according to the minimum safe distance lane change model before this position.Under the premise that the longitudinal travel speed of the vehicle is constant,this paper considers the safety,stability,comfort and other performance indicators of the vehicle.The fifth-order polynomial trajectory planning method is used to design the road convergence track.A trajectory tracking controller based on Model Predictive Control algorithm is established.Combining two-degree-of-freedom vehicle dynamics model and magic tire model we build a vehicle system model and linearize it as the basis of model predictive control;this paper constructs the objective function with the front wheel angle increment and adds the relaxation factor to avoid situations where there is no feasible solution in the solution process.In order to avoid vehicle instability under emergency conditions,we add vehicle dynamics constraints based on front wheel angle and its incremental constraints.Combined with the prediction model,the objective function and the constraint conditions,the optimal solution in the finite time domain is used to realize the rolling optimization in the trajectory tracking control process.The parallel control algorithm is verified by simulation.Based on Prescan/Simulink we builds a simulation environment and selects different working conditions to verify the reliability and effectiveness of the road convergence control algorithm.