Intelligent vehicle transverse and longitudinal comprehensive trajectory tracking method and control system

Abstract

The invention provides an intelligent vehicle transverse and longitudinal comprehensive trajectory tracking method and control system. The method comprises the following steps: building vehicle dynamic models in a longitudinal direction, a transverse direction and a yawing direction, and obtaining a prediction trajectory according to the vehicle dynamic models; determining an objective function and a constraint condition according to the difference between the reference trajectory and the prediction trajectory, solving to obtain an optimal front wheel rotation angle and an optimal wheel slip rate, and establishing a wheel dynamic right inverse model of through a neural network; allowing the wheel dynamic right inverse model to output a wheel torque according to a wheel center longitudinalspeed, a wheel rotating speed and a wheel optimal slip rate which detected by a sensor; and inputting the optimal front wheel rotation angle and the wheel torque into the controlled vehicle, and obtaining a new prediction track through the feedback of the state quantity of the controlled vehicle to achieve closed-loop control. According to the invention, transverse control and longitudinal controlare combined through a tire rotation angle and the slip rate, and the problem of tracking deviation caused by respective control through a decoupling method is solved.

Description

technical field [0001] The invention relates to the field of smart cars or unmanned cars, in particular to a method and a control system for tracking a smart car's horizontal and vertical integrated tracks. Background technique [0002] Existing smart car trajectory tracking research generally decouples the lateral motion and longitudinal motion of the car, and independently designs the lateral controller and longitudinal controller respectively. This control method ignores the horizontal and vertical coupling relationship of smart cars. , For example, when designing the lateral controller, the variables related to the longitudinal motion are regarded as constants, which is obviously inconsistent with the reality. The change of the longitudinal speed will cause some parameters in the lateral control to change, resulting in inaccurate tracking. The more common control method of horizontal and vertical integrated control is to add direct yaw moment control (DYC) at the same ti...

AI Technical Summary

Problems solved by technology

[0002] Existing smart car trajectory tracking research generally decouples the lateral motion and longitudinal motion of the car, and independently designs the lateral controller and longitudinal controller respectively. This control method ignores the horizontal and vertical coupling relationship of smart cars. , for example, when designing the lateral controller, the variables related to the longitudinal motion are regarded as constants, which is obviously inconsistent with the reality. The change of the longitudinal speed will cause some parameters in the lateral control to change, resulting in inaccurate tracking

The more common control method of horizontal and vertical integrated control is to add direct yaw moment control (DYC) at the same time of trajectory tracking. In the process, there will be a phenomenon of tire force redundancy, resulting in tire force not being fully utilized

Images