Global relocation method of surface robot based on millimeter wave radar

Abstract

The invention discloses a global repositioning method for a water surface robot based on a millimeter wave radar, which includes: acquiring the current positioning coordinates C, speed and current environment of the water surface robot point cloud of the millimeter wave radar; and obtaining nearby sensor data according to C to form a query database G; Perform data preprocessing on the millimeter-wave radar point cloud in and G respectively to obtain the current environmental radar point cloud and the environmental millimeter-wave radar point cloud; input and into the environmental feature encoding model to obtain the corresponding feature code f and feature code respectively ; Retrieve the most similar feature code in the corresponding feature code, and obtain the corresponding positioning coordinates from it, as the position identification coordinates of this positioning.

Description

technical field [0001] The invention relates to the technical field of robotics, and more particularly to a global relocation method of a surface robot based on a millimeter-wave radar. Background technique [0002] In recent years, with the rapid development of unmanned technology, the unmanned market has continued to grow. Taking unmanned ships as an example, various surface robots have been gradually promoted to water quality testing, water surface maintenance, bridge testing, water body mapping, maritime search and rescue, etc. In a variety of work scenarios, assist people to work. Positioning is an important part of unmanned driving technology. When the positioning accuracy is low, it is difficult for the robot to obtain its own accurate position, and it cannot accurately reach the designated position and complete the designated task. At present, the global localization algorithm used in the localization of surface robots usually uses lidar data or visual image data to...

AI Technical Summary

Problems solved by technology

Positioning is an important part of unmanned driving technology. When the positioning accuracy is low, it is difficult for the robot to obtain its own accurate position, and it is impossible to accurately reach the specified position and complete the specified task.

At present, the global positioning algorithm used in the positioning of water surface robots usually uses lidar data or visual image data to manually design feature descriptor algorithms, and then completes global positioning through feature matching and prior map regression. However, due to the lack of texture information in lidar data , while the visual information lacks three-dimensional geometric information, and the hand-designed descriptor algorithm leads to low final positioning accuracy

Images