Historical map utilization method based on visual robot

Abstract

The invention discloses a historical map utilization method based on a visual robot. The historical map utilization method comprises the following steps of 1, controlling the visual robot to continuously acquire images of preset road signs in a preset working area and acquire road sign information of the preset road signs; 2, transmitting the acquired road sign information to a first map positioning system and a second map positioning system simultaneously; 3, controlling the first map positioning system and the second map positioning system to process corresponding road sign information so asto obtain first pose information and second pose information; 4, selecting the first pose information and the second pose information under the same preset road sign, and calculating a transformationrelation between the first pose information and the second pose information; and 5, controlling a historical map to perform corresponding transformation operation according to the transformation relation, and using the transformed historical map as a reference for subsequent motion control. By controlling the historical map to be matched with actual physical environment, the method improves the intelligent degree of path planning.

Description

technical field [0001] The invention belongs to the technical field of intelligent control, and in particular relates to a visual robot-based historical map utilization method. Background technique [0002] Every time a traversal is completed in the same working area, the robot often re-builds the map or updates it incrementally. changed to fail. For example, for a sweeping robot, if the map changes, the last unfinished cleaning record will also be invalidated, and tasks such as intermittent sweeping cannot be performed, which reduces the working efficiency of the sweeper. Whenever the work is restarted in the same working area, The physical position and direction of movement of the visual robot are likely to change compared to the end of the last work task, or the current position and the corresponding scene point in the matching map are roughly at the same position, but the visual sensor is in the same position twice. It is impossible to keep the same pose during the acq...

AI Technical Summary

Problems solved by technology

[0002] Every time a traversal is completed in the same working area, the robot often re-builds the map or updates it incrementally. Change and fail

Images