Research On Coordination Control Of Trajectory Tracking And Stability For Distributed Drive Unmanned Vehicle

The "New Four Modernizations" of the automobiles,with "electrification,networking,intelligence and sharing" as the core,has being the development trend of the automobile industry.The dynamics control of driverless vehicles can complete trajectory tracking based on the trajectory output of decision planning and the driving state of the vehicle,and ensure the driving safety and stability.However,the unreasonable planning caused by emergent conditions and poor road conditions can increase the difficulty of trajectory tracking and stability control of unmanned vehicles.In addition,the autonomous steering system fault can seriously affect the trajectory tracking performance of the unmanned vehicle and even make it lose the steering capability.In this paper,the trajectory tracking control and stability control of distributed drive unmanned vehicles are studied to address the above problems.Firstly,the system dynamics model of the distributed drive unmanned vehicle is established with the idea of modular modeling,including the vehicle dynamics model,in-wheel motor model,tire model and autonomous steering system model.In addition,the platform of Matlab/Simulink and Car Sim co-simulation is built to simulate and verify the control strategies.Secondly,a coordinated control strategy is designed for trajectory tracking and lateral stability according to the driving state.The adaptive prediction module is added to the LQR(Linear Quadratic Regulator)trajectory tracking control,and the prediction time is adjusted to make the steering smoother by identifying the steering smoothness based on the dynamics of the tires.Further,the current stability state of the vehicle is determined using the phase plane of side slip angle and side slip angle rate,so as to adaptively adjust the weight ratio of the sliding mode surface in the sliding mode control.The additional yaw moment required for coordinated control can be calculated through the direct yaw moment control,and generated through the tire force directional distribution.The simulation results show that the unmanned vehicle steers more smoothly,and the trajectory tracking and lateral stability are better coordinated.Thirdly,a fault estimation method and fault-tolerant control strategy of autonomous steering system are proposed.A fault model is developed based on steering system fault factors,and the steering system fault value is fusion estimated using Bayesian algorithm.Further,the fault expression is used to reconstruct the trajectory tracking model to dynamically adjust the front wheel steering angle for compensating the trajectory tracking.And then the direct yaw moment control based on the adaptive terminal sliding mode is proposed by combining the cooperative game and terminal sliding mode control theories,and the four-wheel torque distribution is completed for trajectory tracking compensation in differential action under multiple objectives and constraints.The simulation results show that the proposed adaptive terminal sliding mode control has better control effect compared with the conventional sliding mode control when the unknown steering system fault occurs.Finally,a semi-physical hardware-in-the-loop test platform of the autonomous steering system is built,and the coordinated control strategy of trajectory tracking and lateral stability and the fault-tolerant control strategy considering the steering system fault are verified based on this test platform.The test results show that,during the emergency double lane change,the additional yaw moment generated by the stability compensation controller using direct yaw moment control can reduce the yaw rate and side slip angle about 17.73% and 25.8%,respectively,which verifies the trajectory tracking and lateral stability coordination ability of the strategy during the emergency lane change on the medium-low adhesion road.The proposed fault-tolerant control strategy can make the driving trajectory under 0.55 weakening fault close to that when there is no fault.Compared with the fault-free control approach,the maximum lateral deviation and heading deviation are reduced by about 82.92% and 66.66%,respectively,which verifies the fault-tolerant control capability in case of unknown fault inputs.