AGV positioning system and method for fusing 2D environmental map and sparse artificial landmark

Abstract

The present invention discloses an AGV positioning system and method for fusing a 2D environmental map and a sparse artificial landmark. The system comprises a vehicle-mounted device, an artificial landmark, and a 2D environmental map. The 2D environmental map is established by the vehicle-mounted device according to a current environment. The artificial landmark comprises position information andnumber information. Each artificial landmark is uniquely numbered in the environmental map. The vehicle-mounted device comprises a laser radar, a milemeter, a gyroscope, and a positioning unit. The positioning unit comprises a laser data processing module, a sensor data storage module, a point cloud matching graph construction module, an environment positioning module, an artificial landmark positioning module, and a positioning result fusing module. The point cloud matching graph construction module is used for graph construction. After the 2D environmental map is obtained, the environment positioning module and the artificial landmark positioning module are used for obtaining a positioning result. Finally, the positioning result fusing module is used for fusing positioning results and providing final positioning information, so that high-precision positioning of an AGV car in environment is implemented.

Description

technical field [0001] The invention belongs to the field of automatic control, and in particular relates to an AGV positioning system and method for integrating 2D environment maps and sparse artificial landmarks. Background technique [0002] At present, automatic guided vehicles have become important equipment for enterprise internal logistics, smart factories, and advanced logistics, and play a very important role in improving the automation, informatization and intelligence of enterprise internal logistics. The premise of automatic guidance and automatic material handling is that the vehicle can be accurately positioned. The current common laser SLAM positioning is limited by the computing power of the computer and the minimum resolution of the probability map, so the positioning accuracy cannot meet the requirements of some industrial scenarios that require high-precision positioning, and in some environments that are relatively open and beyond the scanning range of th...

AI Technical Summary

Problems solved by technology

The current common laser SLAM positioning is limited by the computing power of the computer and the minimum resolution of the probability map, so the positioning accuracy cannot meet the requirements of some industrial scenarios that require high-precision positioning, and in some environments that are relatively open and beyond the scanning range of the laser radar. The lower laser SLAM cannot complete the positioning that meets the motion control requirements, and the artificial landmark positioning needs to add a large number of artificial landmarks in the environment, which increases the cost of navigation, and when the laser radar of the AGV vehicle is installed at a lower height, the artificial landmarks are easily blocked. This leads to a decrease in positioning accuracy

Images