Vehicle path tracking method based on improved Stanley control

Abstract

The invention discloses a vehicle path tracking method based on improved Stanley control, and the method comprises the steps: carrying out the modeling of a path tracking method according to a Stanley control algorithm based on a two-degree-of-freedom vehicle motion model; an F Stanley algorithm path tracking controller is designed by using the model, a fuzzy controller is designed to continuously adjust and optimize gain parameters of the Stanley controller, transverse control of the unmanned vehicle is realized by improving the controller, and longitudinal control is controlled through a CarSim vehicle model; and performing simulation analysis on the path tracking controller, and evaluating the universality of the improved controller under various trajectories and curvatures. The application range of the vehicle speed is widened, the vehicle path tracking precision and the operation stability are improved, the self-adaptive controller is provided, parameters are automatically adjusted according to the vehicle state and the path course, and the vehicle path tracking accuracy is improved. On the basis of a traditional Stanley controller, gain parameters of the unmanned vehicle trajectory tracking controller are optimized through a fuzzy algorithm, and the path tracking precision is improved.

Description

technical field [0001] The invention relates to the technical field of control systems of unmanned vehicles, in particular to a vehicle path tracking method based on improved Stanley control. Background technique [0002] The control system of unmanned vehicles usually includes a perception layer to collect data around the vehicle and process the collected data; a planning layer to plan the vehicle trajectory; and a control layer to control the vehicle tracking path. Among them, path tracking has been one of the research hotspots in the past decade, and the main task of path tracking is to control the vehicle to travel along a predetermined trajectory. [0003] In path-following controllers, one of the most commonly used strategies is kinematics-based control. Such controllers determine the required steering input based on vehicle kinematics such as acceleration, velocity, and position, which are more readily available than based on dynamics such as forces and torques used ...

AI Technical Summary

Problems solved by technology

[0004] However, further research on the performance of the Stanley controller revealed that the controller is too parameter dependent, which is a common problem with most motion controllers

When the parameters of the Stanley controller are adjusted properly, it will work well in a certain road route and a certain speed range, but when the vehicle speed increases, the tracking lateral error of the Stanley controller will become larger, resulting in a control effect bad question

Images