Path tracking control of an autonomous vehicle with model-free adaptive dynamic programming and RBF neural network disturbance compensation

Abstract

The performance of the model-based controller is always affected by the uncertainty and nonlinearity of the model parameters in the vehicle path tracking process. To address this issue, a novel path tracking controller based on model-free adaptive dynamic programming (ADP) is proposed for autonomous vehicles in this paper. To be specific, the proposed controller obtains information from the online state and front-wheel angle input data which are repeatedly used to calculate the controller gain iteratively. So, this controller features not requiring accurate knowledge of vehicle model parameters for controller development. Meanwhile, the path tracking performance of the autonomous vehicle will be inevitably disturbed by unknown nonlinear external disturbance. To approximate this disturbance, the learning characteristics of Radial Basis Function Neural Network (RBFNN) are applied to generate compensation for the front-wheel angle. Afterward, the weight updating law of RBFNN is derived by Lyapunov function to ensure the stability and convergence of the whole system. Finally, Hardware in the loop (HIL) test results demonstrate that the proposed ADP-RBF controller can improve the comprehensive performance of the vehicle path tracking control system and achieve the balance between path tracking accuracy and minimum sideslip angle.

Designing New Model-Based Adaptive Sliding Mode Controllers for Trajectory Tracking Control of an Unmanned Ground Vehicle

“…is the transformation matrix and ) (q A is the null space of the matrix. The relationship between these two matrices is as shown in (11).…”

“…In practical applications, it is difficult to obtain the dynamic parameters of the system, so controllers that do not require a mathematical model of the system are necessary. Nonlinear control methods that do not require a system model have been used [8], [9], [10], [11] and SMC is a popular choice due to its robustness to uncertainties, parameter changes and external disturbances [12], [13], [14], as well as its good speed response and transient performance [15], [16]. Thanks to these advantages, SMC has often been chosen to perform trajectory tracking control of the UGV [17], [18], [19].…”

ADP-Based Optimal Control for Discrete-Time Systems With Safe Constraints and Disturbances

“…As a result, it becomes difficult to establish an accurate mathematical model, leading to a mismatch between the dynamic model and the controller. This mismatch can have a detrimental impact on control performance [32]. To address external disturbances, Che [33] developed a novel ADP framework to estimate rudder faults and ocean current disturbance by introducing neural network estimators and integrating both rudder faults and external disturbances within the utility function.…”

Research on Nonlinear Trajectory Tracking Control Method for Autonomous Vehicles

“…Currently, trajectory tracking control methods mainly include three main categories: model-free trajectory tracking control methods, linear model-based trajectory tracking control methods and nonlinear model-based trajectory tracking control methods. Model-free trajectory tracking control methods include such as neural network control [1][2], model-free adaptive iterative learning control [3], fuzzy control [3][4], game methods [5], model-free adaptive control [6][7], reinforcement learning [8], and so on. The model-free trajectory tracking control method is a rule-based approach that requires a large amount of experimental data as a basis for design, which makes the design of the controller very difficult.…”