Research On Longitudinal And Lateral Coupling Motion Planning Of Intelligent Vehicle

In the context of the gradual informatization and intelligent development of the trans-portation system,the motion planning of intelligent vehicles reflects important potential values in terms of driving safety,transportation efficiency,energy consumption,and en-vironmental pollution.Motion planning has received widespread attention from national governments and scientific researchers worldwide.The motion planning of intelligent ve-hicles is based on the consideration of vehicle status and road environment information,and takes driving safety,feasibility,and stability as the optimization goals to determine the optimal driving route.Considering the characteristics of the vehicle system;s strong non-linearity and strong coupling,how to provide motion planning for horizontal and vertical coupling is the focus of this paper.This paper is funded by the Key project of the National Natural Science Foundation of China Regional Innovation Development Joint Fund "Research on Key Technologies of Automobile Intelligent Driving Decision and Man-Vehicle Cooperative Control in Ice and Snow Environment"(U19A2069),The National Natural Science Foundation of China In-ternational(Regional)Cooperation and Exchanges Key Project "Safety-oriented Rolling Optimization of Electric Vehicle Energy Efficiency"(No.61520106008),Jilin Province Industrial Innovation Special Fund Project "Research and Development and Industrial-ization of Intelligent Vehicle Active Stabilization Control Unit under Extreme Operating Conditions Based on Predicted Safety"(2019C036-5),and "Thirteenth Five-Year" Science and Technology Project of Jilin Provincial Department of Education "The Vehicle Actively Expands and Stabilizes Rolling Optimization Control under Extreme Operating Condi-tions Based on Predictive Safety"(JJKH20190165KJ),this paper studies the longitudinal and lateral coupling motion planning,mainly including the following aspects:As the basis of the design planning algorithm,the establishment of the vehicle system model is a crucial step.Considering that the vehicle's driving path is determined by the driving direction and driving speed,lateral planning needs to consider the ability of longitudinal planning.Therefore,based on vehicle motion,Newton's law and tire model,a nonlinear vehicle system model combining kinematics and dynamics models is established.Compared with the existing high-precision Hongqi HQ430 series car model in the laboratory,the vehicle dynamics simulation software veDYNA was used for simulation verification to prove the rationality and effectiveness of the established vehicle system model.Aiming at the characteristics of vehicle system nonlinearity and coupling,a trajectory planning strategy based on differential flat longitudinal and lateral coupling under a given path condition is proposed.Selecting the flat output,and the flat output and its finite-order derivatives are used to represent the system's state variables and control inputs.This proves that the established nonlinear vehicle system is a flat system.Combining vehicle driving safety and saturation thresholds of actuators to establish constraints,and driving along a given path,speed,and small planning range as performance indicators,this planning strategy is described as an optimization problem based on differential flatness.Based on this,the state quantity can be directly input to the tracking control module to ensure the safety and stability of the vehicle.For the trajectory planning of the above-mentioned given path,sudden pedestrians,vehicles or fixed obstacles cannot be avoided,which has great limitations.Aiming at the rapidly changing actual traffic environment,a longitudinal and lateral coupling path plan-ning strategy based on model prediction is proposed.Facing the complex and changing driving environment,this paper uses the artificial potential field method to describe the environment,and establishes a road environment model including a virtual gravitation-al field for lane keeping areas and a virtual rectangular repulsive field for obstacle cars.Combining environmental information and vehicle dynamics to establish constraints,and taking obstacle avoidance,lane keeping,tracked expected speed,and small amount of control as optimization targets,this planning strategy is described as an optimization problem based on model predictive control.Because the performance indicators affect each other,a weight coefficient is introduced to adjust the planning demand conflict.Aiming at the proposed controller design method,this paper combines Simulink simulation environment and vehicle dynamics simulation software veDYNA to conduct joint simulation.Based on the optimization toolbox in MATLAB software,it solves differential flat optimization problems and model prediction optimization problems.Proof of the effectiveness and feasibility of the optimization strategy under different working conditions.