Cross-scene navigation method and device and storage medium

Abstract

The invention discloses a cross-scene navigation method and device and a storage medium. The method comprises the following steps: acquiring a current position of a robot and a target position which is to be reached by the robot and is not in the same application scene with the current position; searching the created git navigation map, acquiring a decision chain in the git navigation map corresponding to the current position to the target position, and acquiring a decision route in the application scene corresponding to the decision chain based on the decision chain to obtain a corresponding cross-scene navigation path; and according to the obtained cross-scene navigation path, enabling the robot to navigate from the current position to the target position in a cross-scene manner. The problems that in the prior art, a cross-scene map does not have a navigation error correction function in the application process, and map data management is disordered are solved. Furthermore, the cross-scene navigation efficiency and the accuracy of the navigation route are improved.

Description

technical field [0001] The present invention relates to the technical field of robots, in particular to a cross-scene navigation method, device and storage medium. Background technique [0002] With more and more application scenarios of autonomous mobile robots, the requirements for cross-scenario navigation of autonomous mobile robots are also getting higher and higher. In the navigation process of an autonomous mobile robot, the cross-scenario navigation requirement makes the robot need to read the corresponding map data in different application scenarios, and navigate according to the map data of this scene. Therefore, how to manage and maintain the maps of different scenarios and How to quickly find the map required by the robot is a problem that the robot needs to pay attention to and solve when navigating across scenes. [0003] At present, in the prior art, there are mainly the following two solutions to the above problems. One is: store all the scenes where the rob...

AI Technical Summary

Problems solved by technology

However, this processing method is not convenient for maintaining the map of a single scene

The other is: save a map for each scene of the robot, and go to the scene where the robot goes to find it; the disadvantage is that it cannot reflect the sequence relationship of the robot's decision-making, and there is no error correction function for navigation, such as the robot from the warehouse. When the scene comes out, it should actually go to the factory park scene, but the search result is the home scene. At this time, the robot cannot find this error.

In addition, you can also choose to maintain the current state machine of the robot in the server or robot body software. When the perception moves out of a scene, switch this state and search in the map server in turn to get the corresponding map. The time efficiency is low, and it is impossible to realize the macro-route derivation of robot navigation and the management of navigation map data for the decision-making chain. In addition, it is impossible to display the macro-scene path map

Images