Mobile robot map construction method, storage medium and device based on closed-loop detection and correction

Abstract

The invention discloses a mobile robot map construction method, storage medium and equipment based on closed-loop detection and correction. The method includes step S1, constructing a memory perception template representing a spatial position according to the environment information collected by a visual system, and generating a corresponding visual perception memory library; Step S2: Update the position perception information according to the self-perceived motion information collected by the gyroscope and accelerometer; Step S3, associate the current memory perception template with the visual perception memory library, and calculate the memory perception template when arriving at the experienced environmental position The activity intensity activates the corresponding visual perception memory bank; step S4, draws and corrects the self-perception space map after interconnecting the obtained position perception information with the visual perception memory bank and the memory perception template. The invention realizes real-time updating of position information, eliminates accumulated errors, and improves the accuracy of closed-loop detection in consideration of the excitability level difference between adjacent memory perception templates.

Description

technical field [0001] The invention belongs to the technical field of Simultaneous Location And Mapping (SLAM), and relates to a mobile robot map construction method, storage medium and equipment based on closed-loop detection and correction. Background technique [0002] The core problem of simultaneous positioning and map creation (Simultaneous Location And Mapping, SLAM) is that the robot is required to first explore the environment in an unfamiliar environment to understand the environment (build a map), and use the map to track the position of the robot in the environment simultaneously. (position). Traditional solutions to SLAM problems are mainly based on mathematical probability methods, among which Kalman filter, particle filter and maximum expectation algorithm are the basic solutions to robot SLAM problems. Although these traditional SLAM algorithms still use laser ranging and sonar ranging to collect information, the information collected by these sensors is of...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a mobile robot map construction method, storage medium and equipment based on closed-loop detection and correction to solve the problem of poor image matching accuracy and poor positioning accuracy due to the fact that the position coordinates cannot be updated and corrected in real time in the prior art. Technical problems such as the serious distortion of the two-dimensional position estimation map

Images