| abstract |
The invention discloses a SLAM method based on integration of a laser radar and a rotary inertial navigation framework and information fusion, which comprises the following steps: and a characteristic extraction step, namely calculating translation and rotation between laser frames through IMU data, eliminating motion distortion in the original point cloud, and extracting stable edge and plane characteristics from the original point cloud. And LIO, which consists of a state recursion submodule and an updating submodule, performs iterative Kalman filtering, and outputs initial mileage and undistorted characteristics. And a map building step, namely improving the initial mileage by using the global map, outputting a new mileage and then updating the map by using the new characteristic. The method provided by the invention can be applied to the MEMS inertial navigation with medium and low precision, and effectively improves the positioning precision of the inertial navigation system during long-time operation, thereby improving the accuracy and reliability of map construction. |