3D obstacle avoidance indoor robot navigation method and system

Abstract

The invention discloses a 3D obstacle avoidance indoor robot navigation method and system. The method comprises acquiring a 3D model, an initial position and a final position of a robot, acquiring a global 3D map, carrying out global route planning to produce a series of global path points, carrying out 3D obstacle collision detection on each one of the global path points to obtain the optimal global route plan, making the robot travel according to the optimal global route plan, carrying out 3D obstacle collision detection in the next global path point in travelling, if the robot produces impact, carrying out global route planning through a real-time 3D map as the global 3D map and a real-time position as an initial position, and if the robot does not produce impact, making the robot travel to the final position so that robot autonomous navigation and 3D obstacle avoidance are realized. The method and system can successfully acquire the feasible robot path in the crowded indoor environment, and completes the autonomous navigation and the 3D obstacle avoidance task of the robot from the initial position to the final position.

Description

technical field [0001] The invention belongs to the field of computer vision, and more specifically relates to a three-dimensional obstacle avoidance indoor robot navigation method and system. Background technique [0002] In recent years, indoor mobile robots have attracted more and more attention from researchers because of their wide range of uses. Indoor mobile robot is a comprehensive system integrating multiple functions such as environment perception, dynamic decision-making and planning, behavior control and execution. A key issue in the navigation process of mobile robots is whether the dynamic obstacle avoidance and global navigation tasks can be completed in real time. At present, the more popular obstacle avoidance method is to use lidar or inertial measurement elements combined with visual sensors to complete this task, but the cost is relatively high. With the popularity of cameras and the development of computer vision theory, using pure vision-based methods...

AI Technical Summary

Problems solved by technology

However, due to the increasingly complex indoor scenes, real-time obstacles are easy to appear in the scene, but due to the limitation of robot computing resources, it is difficult for the robot to directly deal with real-time obstacles while navigating. Dynamic obstacles, it is difficult to ensure the accuracy of information acquisition when there are a large number of obstacles

Images