A real-time SLAM scene map constructing system, a navigation system and a method

Abstract

A real-time SLAM scene map constructing system, a navigation system and a method are disclosed. The system overcomes a difficult problem of real-time map constructing for SLAM navigation systems whenscene changes are large, such that a robot can obtain accurate positioning information in a changed scene. The real-time SLAM scene map constructing system includes at least two independent SLAM mapping subsystems having fixed spatial locations. The SLAM mapping subsystems are configured to send spatial geographical location information and collected local scene map information into which a timestamp is added to a core server. The core server is configured to determine the relative location relationship among local scene maps having same timestamps according to the spatial geographical location information of the SLAM mapping subsystems, thus fusing the local scene maps having same timestamps to generate a real-time overall SLAM scene map.

Description

technical field [0001] The invention belongs to the technical field of navigation, and in particular relates to a real-time SLAM scene map construction system, a navigation system and a method. Background technique [0002] Simultaneous Localization and Mapping (SLAM) technology refers to placing the robot in an unknown environment, starting from an unknown position to create an incremental map of the environment, and using the created map for autonomous positioning and mapping. navigation. SLAM technology can enable robots to achieve true autonomous navigation. Positioning algorithms and map representations are usually applied in pairs. In the SLAM problem, robot positioning depends on the environment map, and the map creation process also depends on the accurate positioning of the robot. This is a highly related process that must be solved at the same time. Neither problem can Solved individually. [0003] The SLAM problem has been developed for nearly three decades sin...

AI Technical Summary

Problems solved by technology

When the scene is relatively large, rebuilding the environment map will be a very troublesome thing

[0008] (2) In some specific scenarios, it is necessary to use a laser radar or visual sensor to build a large environmental map during the movement process, which often requires high performance of the sensor and airborne resources itself, and low-cost radar is often difficult to perform the mapping task , the cost of radar used in industrial scenarios is higher

And for some scenes that cannot be closed-loop, such as long corridors, due to the existence of cumulative errors, there are deviations in the SLAM map, which eventually leads to large navigation and positioning errors and affects the accuracy of positioning.

Images