Research On Path Tracking Control Of Vehicle Based On The Look-ahead Scheme

With the steady increase in the number of cars, a variety of social problems causedby the car also appear. The number of deaths of our country due to car accidents isseveral hundreds of thousands every year, most of these car accidents are caused byoperating errors; in addition, traffic jam caused by the unreasonable path planning is aserious problem, it is considered to be an efficient path to solve these problems thoughdeveloping unmanned vehicles. At present, the study on path tracking control ofunmanned vehicles is few, the existing researches usually focus on the position errorcontrol, it is rare to improve driving stability through controlling sideslip angle, thecontrol algorithm for the case of parameter uncertainty also requires further study at thesame time.The path tracking control algorithm of unmanned four-wheel steering vehicle isstudied in this paper, a nonlinear tire model is established in this paper bases on the"Magic Formula", the dynamical equations of four-wheel steering vehicle and the errormodel base on the geometric look-ahead scheme are established at the same time.The path tracking controller and the longitudinal distance controller are built baseon sliding mode control and the effect is validated though the simulation. Theimportance of coupling design and analysis to the research on vehicle control system ishighlighted though comparing the results of coupling and uncoupling simulation. Byanalyzing and comparing the tracking errors of front wheel steering and four-wheelsteering vehicle under the condition of the same road curvature, the advantages of thefour-wheel steering vehicle on operating stability and path tracking capability aredisplayed. In addition, the results of the simulation on three roads which have differentcurvatures show that the four-wheel steering vehicle has a stronger adaptability to theroad with different curvature.An amendment algorithm of road curvature bases on the look-ahead scheme isproposed so that the operating stability and comfort of vehicle will not decline becauseof the road curvature mutation, the effect is verified in the simulation. A sliding modeadaptive controller is designed to work when the vehicle parameters are constant butunknown, it can be founded that the controller can control the deviations and estimatethe unknown parameters accurately.A fuzzy controller is designed for the case that the vehicle model is unknown, andthe joint simulation in this paper is bases on Carsim and Matlab. The path trackingcontrol simulation results of unmanned front wheel steering vehicle and four-wheelsteering vehicle are compared in the paper. The effect of fuzzy controller is validated which can control a vehicle tracking the target path when the vehicle model is unknown.