Unmanned vehicle path planning and tracking control method based on differential flatness and active-disturbance rejection

Abstract

The invention provides an unmanned vehicle path planning and tracking control method based on differential flatness and active-disturbance rejection and can improve the tracking effect of a four-wheel steering unmanned automobile during high-speed overtaking. The method is advantaged in that step 1, a three-degree-of-freedom four-wheel steering automobile monorail control model is established; step 2, based on the control model established in the step 1, according to the differential flatness theory, an under-actuated controlled model is converted into an input output coupling model having disturbance and having no zero dynamic subsystem; step 3, a path planning layer is established above a tracking control layer; and step 4, according to the input output coupling model based on the step 2, an active-disturbance rejection controller based on a generalized proportion integration observer is designed, and tracking of locus planned in the step 3 is realized.

Description

technical field [0001] The invention belongs to the field of path planning and tracking control of four-wheel steering vehicles, and relates to a path planning and tracking control method for unmanned vehicles based on differential flatness and active disturbance rejection. Background technique [0002] The four-wheel steering system (all-wheel steering system) is an important part of vehicle handling stability. It is characterized by the front and rear wheels turning in opposite phases at low speeds to reduce the turning radius; turning in the same phase at high speeds makes it easier to change lanes. Under the premise of continuous improvement of vehicle handling stability and vehicle intelligence requirements, the requirements for accurate trajectory planning and tracking control of reference trajectory (path planning) for four-wheel steering vehicles are becoming more and more urgent. [0003] Differential flatness theory is an effective way to solve the problem of under...

AI Technical Summary

Problems solved by technology

The MATLAB simulation was carried out under the condition that the initial error of the model was large and there were internal disturbances (small angle omitted items) and external disturbances (longitudinal wind resistance and lateral gusts). The simulation results showed that the designed method achieved a good tracking effect, linear The ADRC (using a high-order extended state observer) and the generalized proportional integrator have basically the same tracking effect. In the case of particularly complex disturbances, the third-order polynomial approximation effect of the generalized proportional-integral observer may not be as good as the high-order extended state observer

Images