Robot path robustness planning method based on monocular camera ranging uncertainties

Abstract

The invention discloses a robot path robustness planning method based on monocular camera ranging uncertainties. The method comprises the following steps of 1, on the basis of an existing pixel pointenvironment model, segmenting a path in a ligature direction of a path starting point and a target point at equal intervals to form a whole path; 2, constructing an uncertainty optimization model, wherein the process that the length of the path is the shortest is taken as a target function, and the process that a robot does not collide with obstacles is taken as a constraint condition; 3, establishing a luminance function aiming at the path planning uncertainty optimization model, utilizing a firefly intelligent algorithm for solving a path planning problem based on the luminance function, andoutputting an optimal path. According to the robot path robustness planning method, the luminance function is established aiming at the path planning uncertainty optimization model, two indexes of path length and path safety degree are considered, and in this way, the method has a good obstacle avoidance function and can plane a short path.

Description

technical field [0001] The invention relates to a robot path planning method, in particular to a robot path robustness planning method based on monocular camera ranging uncertainties. Background technique [0002] Path planning refers to finding the most reasonable and efficient obstacle-avoiding path from the starting point to the target point in an environment distributed with obstacles. Path planning is one of the key technologies in mobile robot navigation. Existing path planning methods are divided into classic path planning and intelligent path planning. Classical path planning methods include: artificial potential field method, probability map method, and fast search random tree method. Classical path planning methods sometimes cannot generate optimal paths, and often is locked in a local minimum. Some of them may not provide a suitable solution in the presence of multiple obstacles or in a dynamic environment. [0003] Therefore, classical path planning has been gr...

AI Technical Summary

Problems solved by technology

[0006] The invention patent with application number 201710019066.9 discloses a feedback path planning method in the presence of measurement uncertainty. By considering the influence of measurement uncertainty on the feedback field of attraction, the Pondriakin set difference is used to narrow the field of attraction. size, re-evaluate the problem of safe obstacle avoidance in the path planning process, and use the GP algorithm to estimate the size of the attraction domain online to overcome the defect of slow calculation of the attraction domain

Images