Path Tracking Control Of Unmanned Sightseeing Vehicle In Open Parks

Unmanned driving technology is one of the technologies that are currently attracting attention.Its application can not only effectively solve the problem of traffic congestion,but also improve traffic safety and promote the upgrading of the transportation industry.Among the many technologies of unmanned vehicles,path planning technology and path tracking technology are the core technologies of unmanned vehicles.For unmanned vehicles in specific scenarios,such as unmanned express vehicles and unmanned sightseeing vehicles,they will enter the market earlier due to their relatively simple driving environment and huge market demand.Therefore,this paper aims to promote the further development of unmanned driving technology in practical applications by studying the local path planning and path tracking technologies of unmanned sightseeing vehicles in open parks.Aiming at the problems of complex semi-structured road conditions and difficult path planning in the park,a robust obstacle avoidance path planning method with unknown obstacle distribution is proposed.Firstly,an improved vector field histogram(VFH)algorithm is proposed to determine the optimal passable area,and then a method for determining the target state is given based on the optimal passable area;secondly,according to the starting point and target state,a description method of obstacle avoidance paths based on piecewise quadratic Bezier curve is proposed.Then,the optimization problem of the curve parameters is established based on the curvature constraints of the vehicle operation and the direction change margin constraints of the target point,and the curve parameters are optimized by using the sequential quadratic programming algorithm;finally,the proposed method is verified by simulation and real vehicle experiments.Compared with other methods,this method has the shortest path and better robustness.A vehicle motion model based on the mechanism model and data-driven is proposed to solve the problem that there are errors between the mechanistic model and the unmanned sightseeing vehicle and the data-driven modeling is affected by the fluctuation of the collected data.The vehicle mechanism model is used as the prior knowledge of the data-driven model,and the state predicted by the mechanism model is compared with the actual data to obtain the error between the mechanism model and the actual vehicle.Then,the Auto-Regressive with Extra Inputs(ARX)model is used to compensate the error to obtain more accurate vehicle motion state prediction results.In addition,this chapter improves the parameter identification method of the ARX model,and limits the weight of the parameter change in the performance index function of the commonly used least squares method.Finally,the accuracy and stability of the model are verified by numerical simulation through Matlab.Aiming at the problem of complex vehicle parameters and sensor data fluctuations in the path tracking process of unmanned sightseeing vehicles,a path tracking adaptive control algorithm based on mechanistic and data-driven vehicle motion model was is proposed.The algorithm estimates and adjusts the system parameters through an adaptive controller to achieve control of path tracking.This chapter introduces the basic concept and design methods of adaptive control,and the method of model-free adaptive control(MFAC).Finally,the adaptive control scheme of path tracking designed in this chapter and MFAC are compared and analyzed through simulation experiments,which is verified that the controller designed in this paper has fast response ability and better antiinterference ability.