Sweeping robot obstacle recognition and avoidance method based on depth vision

Abstract

The invention provides a sweeping robot obstacle recognition and avoidance method based on depth vision, which is used for enabling a sweeping robot to sense obstacles in a front working area in advance and adopt a corresponding obstacle avoidance mode according to the types of the obstacles. The sweeping robot obstacle recognition and avoidance method based on depth vision comprises the following six steps: firstly, carrying out information acquisition on an area in front of the sweeping robot by using a depth camera; then extracting spatial position information of each pixel point in the depth image, and marking a ground area; marking a path space region in the depth image, and segmenting an image of the obstacle; according to the obstacle image, using an x-axis and y-axis pixel distribution histogram and a depth histogram to distinguish the obstacle category; selecting a detour obstacle avoidance mode or a turn-back obstacle avoidance mode according to the obstacle category, and executing an obstacle avoidance operation; and finally, enabling the sweeping robot to run forwards and carry out normal sweeping, and executing the first five steps circularly.

Description

technical field [0001] The invention relates to the field of robots, in particular to a depth vision-based obstacle recognition and obstacle avoidance method for a sweeping robot. Background technique [0002] Nowadays, sweeping robots have gradually entered the public's field of vision, and their advantages of being able to easily clean up dust, hair, paper scraps and other garbage without manual operation are deeply loved by the majority of people. However, my country's current research on sweeping robots is still in its infancy. Most of the existing obstacle avoidance methods of sweeping robots use passive response sensors such as lasers, ultrasonic waves, and pressure. These sensors often do not have the ability to judge path obstacles. Therefore, there is often the risk of getting stuck between the dining table or dining chairs, getting lost under the dining table, and being easily entangled by small obstacles such as plastic bags, indoor wires, and hangers. [0003] Th...

AI Technical Summary

Problems solved by technology

However, my country's current research on sweeping robots is still in its infancy. Most of the existing obstacle avoidance methods of sweeping robots use passive response sensors such as lasers, ultrasonic waves, and pressure. These sensors often do not have t

Images