Point cloud alignment method and device, equipment and storage medium

Abstract

The invention discloses a point cloud alignment method and device, equipment and a storage medium, and belongs to the field of high-precision maps. The method comprises the following steps: acquiring a first point cloud and a second point cloud; calibrating at least one homonymy point pair in the first point cloud and the second point cloud based on the calibration operation, wherein the homonymy point pair comprises a first point in the first point cloud and a second point in the second point cloud; optimizing the first acquisition pose and the second acquisition pose according to the position difference between the first point and the second point in the at least one homonymy point pair; and performing point cloud alignment on the first point cloud and the second point cloud according to the optimized first acquisition pose and the optimized second optimization pose. In the point cloud alignment process, whether the point cloud is a rigid body or not does not need to be considered. In addition, whether the object in the environment for collecting the point cloud moves or not does not need to be considered. The accuracy of a point cloud alignment result can be improved.

Description

technical field [0001] The present application relates to the field of high-precision maps, and in particular to a point cloud alignment method, device, equipment and storage medium. Background technique [0002] In the process of building a high-precision map, it is necessary to collect the point cloud corresponding to the road through the collection device, for example, deploy the collection device on the vehicle, and collect the point cloud corresponding to the road while the vehicle is driving on the road. Since each collection can only collect point clouds corresponding to some areas in the road, it is necessary to merge the point clouds collected multiple times. In the process of merging point clouds, it is necessary to use the collection pose of the collection device when collecting point clouds. Due to the error in the acquisition pose of the measured acquisition equipment and the time error at different acquisition times, it may appear that the points collected at ...

AI Technical Summary

Problems solved by technology

[0004] The premise of the ICP algorithm is that the point cloud in the area is a rigid body, but in the process of collecting the point cloud by the acquisition device, complex phenomena such as multiple refraction and reflection will occur in the real environment, and the assumption that the point cloud in the area is a rigid body is not always true. of

In the case of non-rigid body point clouds, the result of point cloud alignment using the ICP algorithm will be affected by the non-rigid body point cloud, resulting in poor point cloud alignment.

The accuracy of the result of point cloud alignment by the above method is low

Images