A new vehicle path-following strategy of the steering driver model using general predictive control method

Abstract

In order to build an accurate and effective model, simulation of driver behavior is absolutely essential for the development of advanced driver assistance systems and the current assessment of vehicle handling stability. The purpose of the proposed active steering control is to assist the driver to follow the desired path, especially in situations of the vehicle under strong external disturbance, distracted driver, or other unforeseen circumstances that can cause deviations. Based on the preview optimal simple artificial neural network driver model, an active steering system using general predictive control method is established. In order to improve the path-following capability of vehicles under disturbances, a general predictive controller, with the deviation between the vehicles’ actual and desired lateral positons as inputs and with the corrective steering wheel angle as outputs, is developed to follow the desired path. Meanwhile, adapting recursion least square method with the forgetting factor to estimate the parameters of the controlled auto-regressive and integrated moving average model of the vehicle is designed. The proposed vehicle path-following control system is evaluated in some typical conditions (e.g. under strong crosswind condition in standard double-lane-change, etc.). Simulation results and analysis have verified that this new vehicle path-following strategy, given by general predictive controller, is capable of capturing driver’s steering behavior and the amended driver steering angle can improve the dynamic performance of vehicle under some external disturbances.

Implementation of Human-Like Driver Model Based on Recurrent Neural Networks

“…Lin et al [26] proposed a ANN-based driver handling behaviour model with the position, velocity and acceleration of the vehicle used as the input and the front wheel steer angle considered as the desired output of the trained network. Caoet et al [27] developed a new vehicle path-following strategy based on the preview optimal simple artificial neural network (POSANN). Model predictive control (MPC) is an efficient trajectory tracking control algorithm and is often used to build driver model [28].…”

Model predictive control–based steering control algorithm for steering efficiency of a human driver in all-terrain cranes

“…Information about the state error (i.e. the lateral and yaw angle errors), as shown in Figure 5 and equations (7) and (8), is necessary for the design of a steering control algorithm.…”

Lane‐keeping control based on an improved artificial potential method and coordination of steering/braking systems

“…Based on the deviation, a PD controller was designed to control the lateral motion of the vehicle, so that the vehicle could track any complex path. In [10], the preview optimal simple artificial neural network driver model is designed. Although the preview optimal driver model is more accurate and sensitive, its control algorithm is too complex.…”