Multi-sensor map construction method and device based on visual SLAM and chip

Abstract

The invention relates to a multi-sensor map construction method and device based on visual SLAM and a chip. The distance to a target object is obtained according to an initial detection signal transmitting time and an echo detection signal receiving time of a laser radar sensor, the orientation of the target object is obtained according to an initial detection signal and an echo detection signal,first point cloud information is obtained according to the distance to the target object and the orientation of the target object, the information matching is performed on the first point cloud information and initial point cloud information based on the distance to the target object and the orientation of the target object, and the first point cloud information and the initial point cloud information are fused according to a matching result to construct a point cloud map such that the obtained point cloud information is fused with original point cloud information. The precision of the formedmap is high such that a robot can conveniently realize autonomous obstacle avoidance and route planning.

Description

technical field [0001] The invention relates to the technical field of intelligent robots, in particular to a method, device and chip for constructing multi-sensor maps based on visual SLAM. Background technique [0002] Generally, although the direct map constructed by the feature point map has a small amount of data and is convenient for data storage and extraction, it often cannot reflect some necessary information of the environment, such as the location of obstacles in the environment, so this type of map can basically only It is used for positioning problems and cannot allow robots to perform autonomous barrier and path planning. Moreover, when existing robots construct maps, most of them use visual sensors or laser sensors alone, and the acquired data is valid, but the map accuracy is low. Contents of the invention [0003] The present invention provides a method, device and chip for constructing a multi-sensor map based on visual SLAM, the main purpose of which is...

AI Technical Summary

Problems solved by technology

[0002] Generally, although the direct map constructed by the feature point map has a small amount of data and is convenient for data storage and extraction, it often cannot reflect some necessary information of the environment, such as the location of obstacles in the environment, so this type of map can basically only It is used for localization problems, and it is impossible for the robot to perform autonomous obstacle and path planning

Moreover, most existing robots use visual sensors or laser sensors alone when constructing maps, and the acquired data is valid, but the map accuracy is low

Images