Lateral And Longitudinal Control Of Unmanned Vehicles

The Longitudinal and Lateral control systems of unmanned vehicles is one of the core parts to ensure the safe driving of unmanned vehicles.Unlike the manned vehicles,unmanned vehicles need more advanced Longitudinal and Lateral control algorithms to ensure that the vehicles can accurately track the desired signals given by decision making planning module.In view of the above problem,a Longitudinal and Lateral control system of unmanned vehicles is designed and implemented in this thesis.The main work and innovation points are as follows:First,In the past,most researchers used simplified or partial modeling methods to model the whole vehicle.Different from this method,in this thesis,the whole vehicle model in Carsim is taken as the controlled object,which is more suitable for the real running state of the vehicle,and improves the reliability of the Longitudinal and Lateral controllers.Second,Active disturbance rejection control(ADRC)is applied to the lateral control of unmanned vehicle to solve the path tracking problem of unmanned vehicle in the case of lateral and longitudinal coupling.The ADRC algorithm does not depend on the precise mathematical model.The algorithm of ADRC does not depend on the precise mathematical model,and the path tracking control of the unmanned vehicle is realized by controlling the wheel angle of the vehicle.Three typical conditions(double line change condition,single line change condition and position tracking under the speed change condition)are set to verify the rationality of the lateral controller.Comparing with the simulation results of traditional PID control algorithm,it shows that the designed lateral controller can accurately track the lateral position,and can still achieve good tracking effect in the case of the speed change;it shows that the ADRC algorithm can effectively solve the lateral position tracking problem in the case of vehicle Longitudinal and Lateral coupling.Third,Using the idea of hierarchical control,a longitudinal control scheme is designed:(1)In the upper control of the longitudinal control system,the Radial Basis Function(RBF)neural network is combined with the traditional PID control algorithm to achieve the optimization of traditional PID control;(2)In the lower control system,the expected acceleration from the upper control is received.The throttle of the driving control system or the brake pressure of the brake control system is calculated by the inverse longitudinal dynamic model.At the same time,a reasonable switching control strategy is designed to avoid frequent switching between the two systems.Finally,three simulation conditions(sinusoidal condition,complex condition,speed control condition under direction interference)are selected,and the longitudinal speed tracking results are compared with the traditional PID control algorithm.The simulation results show that the longitudinal speed tracking controller can improve the control accuracy of speed tracking,and the control accuracy is not affected when the direction changes.Finally,based on the longitudinal speed tracking control and considering the influence of the front vehicle on the longitudinal control system,the longitudinal control considering the distance tracking pattern is proposed.Three problems are mainly solved in this thesis: the choice of expected distance,the choice of distance control algorithm and the switch strategy of setting speed tracking pattern and distance tracking pattern.The integral separation PID is used as the distance tracking control algorithm to track the vehicle distance.Considering the influence of noise and the limitation of equipment,the strategy of Constant Time Headway is adopted.Considering the comfort,security and legitimacy of switching,a reasonable switch strategy is designed.In the last,the rationality of the longitudinal controller considering the distance tracking pattern is verified by setting a variety of typical simulation conditions(steady-state following condition,front vehicle rapid deceleration condition,front vehicle cut in simulation condition,front vehicle cut out simulation condition).Simulation results show that the proposed algorithm can ensure the comfort and safety of car following.