Path tracking method based on combination of Pure Pursuit with PI with genetic algorithm for parameter setting

Abstract

The invention relates to the technical field of path tracking for automatic driving, more particularly to a path tracking method based on combination of Pure Pursuit with PI with a genetic algorithm for parameter setting. On the basis of combination of the Pure Pursuit algorithm with the PI controller, the PI controller corrects a micro horizontal error and parameters of the PI controller are setbased on the genetic algorithm, so that the driverless car tracks the path rapidly and accurately. According to the traditional path tracking algorithm, a target path is tracked only based on the PurePursuit algorithm; the tracking effect of the algorithm depends entirely on a preview distance; a side shearing problem is caused by the too long preview distance, a problem of too serious shaking ofthe turning angle is caused by the too short preview distance, and a problem of adjusting the preview distance difficulty is also caused. Therefore, on the basis of combination of the PI controller with the Pure Pursuit algorithm, the sensitivity of the control accuracy to the preview distance can be reduced; and the error correction effect by the PI controller is improved by setting the PI parameters based on the genetic algorithm.

Description

technical field [0001] The present invention relates to the technical field of path tracking of automatic driving, and more specifically, to a path tracking method based on Genetic Algorithm-based Pure Pursuit and PI. Background technique [0002] In the modern field of autonomous driving, path tracking is an essential technical part. To achieve autonomous driving, unmanned vehicles need to be able to accurately track the target path in order to be able to navigate accurately. Just like people walking, they must be able to control their legs to walk on the target path in order to walk precisely according to the established route. There are many kinds of existing path tracking algorithms, such as Pure Pursuit, Stanley, Alice method, and PID controller. [0003] Each path tracking method has its own advantages and disadvantages. For example, the effect of the Pure Pursuit method largely depends on the selection of the preview distance. If the preview distance is too long, it...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome at least one defect described in the above-mentioned prior art, the present invention provides a path tracking method based on the Pure Pursuit of genetic algorithm tuning parameters combined with PI, which utilizes the advantages of Pure Pursuit algorithm and PI controller complementarily, and uses genetic algorithm Solve the problem of difficult parameter setting of PI controller, and get a path tracking algorithm with good robustness, which can accurately track the established path

Images