Research On Path Tracking Control Of Unmanned Vehicles With Local Path Planning

With the improvement of industrial computing speed and sensor technology,driverless vehicles are the inevitable choice for future transportation.The traditional mode of automobile not only brings traffic congestion,environmental pollution and travel safety and other problems,but also consumes a lot of manpower.Therefore,the research of energy saving,environmental protection and safe and stable driverless vehicles has become a top priority.As one of the important technologies in the field of unmanned driving,artificial intelligence needs to complete complex functions such as real-time positioning,environment perception,path planning and tracking control of vehicles.In the face of obstacles,replanning the global path to avoid obstacles and driving according to the reference path is the key to realize driverless driving.Based on model predictive control algorithm,path tracking control and local path planning of unmanned vehicles are studied in this paper.Aiming at the problem that the tracking accuracy of traditional model predictive control algorithm is not high enough in path tracking of unmanned vehicles,a model predictive control strategy with nonlinear compensation is proposed.Firstly,the vehicle dynamics model is constructed based on Newtonian mechanics,and the vehicle tire model is further processed and simplified,so as to obtain the nonlinear vehicle model for trajectory tracking.Then,after linearizing the vehicle model,nonlinear compensation items are added to improve the prediction accuracy of the whole algorithm for the future state in the prediction stage,so that the algorithm can solve the optimal control quantity which is more in line with the current vehicle state,and realize the accurate tracking of the desired trajectory of the vehicle.Finally,the simulation results show that the tracking accuracy of the system is improved by13% and 60%,respectively,when the compensation is introduced into the prediction model at low and medium speeds.In addition,when the vehicle is in high speed,the controller still has the problem of insufficient precision after introducing nonlinear compensation.The influence of tracking controller parameters on the control performance was studied through theoretical analysis.Based on the effectiveness of the parameters to improve the control performance,the sampling period was selected as the parameter to be adjusted.The fuzzy controller was designed to realize that the sampling period of the tracking controller could be adjusted adaptively according to different speed and road curvature during vehicle running.Through simulation comparison,it is found that the tracking accuracy of the vehicle is further improved by 68% under the condition of high speed.In the aspect of local path planning,in order to solve the problem of excessive obstacle avoidance when the vehicle faces the obstacle environment.Based on the vehicle point mass dynamics model,a new obstacle avoidance function is designed by analyzing the relationship between the vehicle and the position state of obstacles.The obstacle avoidance function makes full use of the state information of the vehicle,so as to avoid the excessive obstacle avoidance phenomenon in the reference path planned by the model predictive control planner.The simulation results show that compared with the original obstacle avoidance function,the designed obstacle avoidance function has better smoothness in trajectory planning.