System Identification And Motion Control Of Skid-steered Model Vehicle

In recent years,with the rise of popular applications such as unmanned driving and smart warehousing,more and more mobile carriers adopt wheel-type skidding to achieve steering.The mobile carrier adopting this steering mode has simple steering actuator,high carrier space utilization,and strong low-speed maneuverability.However,this structure also has the disad-vantages of complicated steering mechanism,greater resistance to be overcome during steering,and difficulty in coordinated control.Generally,its motion module is a non-linear system com-posed of key components such as the chassis,tires,and engine,and requires the cooperation of multiple components to achieve smooth motion.It uses the difference in the speed of the tires on both sides to make the car body slip to achieve steering.However,the interaction between the tire and the ground during steering is more complicated,which makes it difficult to accu-rately model the system and also increases the difficulty of the controller design.Therefore,in order to reduce the difficulty of system modeling and controller design,the research on identifi-cation modeling and control design of sliding steering mobile carrier system has very important application value.In order to study the modeling and control of the slip steering structure,this paper builds a four-wheel model car used to complete the investigation of the slip steering dynamics mech-anism,system identification and motion control research.First,this article introduces the me-chanical structure used by the wheeled skid steering model car and main circuit hardware,and then briefly introduces the software framework and its operating logic.Then,in view of the non-linear system characteristics of the slip steering model car,this paper starts from the vehi-cle dynamics theory and deeply studies the steering mechanism of the slip steering model car.It focuses on the solution of vehicle dynamics model and tire force,derives the nonlinear state space equation of the system in a steady state of motion,and conducts in-depth analysis and verification of the state space equation model based on the experimental platform.Then,two types of system models are proposed and analyzed in depth to lay the foundation for system model identification.The first type of configuration introduces the concept of oper-ating point.The system is approximately linearized near the operating point,and the system can be described by a set of approximately linear models.The second category divides the system into a nonlinear part and a linear part according to the mechanism model and the Hammerstein model idea.In the fourth chapter,this paper studies the model identification schemes of the two type of system.For the first type,the identification structure of the linear model is determined according to the mechanism model of the wheeled skid-steering model car.Then,the Hankel matrix system model identification algorithm based on impulse response was used to identify the model of the actual system,and the transfer function model of the dual-input dual-output system near the operating point was obtained.The experimental data and the results of mecha-nism modeling verify the correctness of the system identification method and the effectiveness of the model.For the second type,a DC compensation scheme based on the mechanism model is determined,so that the object can be approximated as a linear model after compensation,which reduces the adverse effects of low-frequency nonlinear factors on the control of slip steering vehicles.Then,the re-identification experiment is carried out on the controlled object which is approximate to the linear model after compensation,and the results of the two types of models are compared and summarized at the end.Finally,in order to ensure that the system can meet the multi-objective control performance indicators such as rapidity,robustness and anti-disturbance after the system is closed-loop,the design methods of modern control theory can be applied in the control design.According to the re-identification model set,this paper firstly designs a feedforward dynamic decoupling con-troller for the wheeled slip steering model car to suppress the mid- and high-frequency interfer-ence of the coupling item.based on the identification model set to describe the entire system with uncertainty,and based on the extended internal model LQR control algorithm,the robust stability constraint and the anti-disturbance performance index are proposed,and the motion controller is designed under the constraints of multiple control performance indexes.Finally,the actual verification is carried out on the actual platform and compared with the simulation results.The results show that the designed controller has achieved good results.