Instant dense reconstruction method for space target

Abstract

The invention provides an instant dense reconstruction method for a space target, which comprises the following steps: S1, acquiring an RGB-D image of the space target through a depth camera, converting a depth map of the image into a corresponding point cloud, obtaining a three-dimensional coordinate and a normal vector of each point in the point cloud, initializing the pose of an adjacent depth map by using an iterative nearest point algorithm, and obtaining a three-dimensional coordinate of each point in the point cloud; surface elements are used for representing a dense model of the space target; s2, by taking geometric error/luminosity consistency as a constraint condition, registering the depth map/color map of the current frame with the depth map/color map of the previous frame predicted by the dense model, and estimating the camera attitude at the current moment; s3, encoding the current frame, comparing the current frame with the encoding value of the previous frame stored in the database, and judging whether to update the key frame; and judging whether a closed loop exists in the rolling motion of the space target or not, if so, registering the current frame to the most similar key frame, and optimizing the dense model by using the deformation graph to eliminate accumulative errors.

Description

technical field [0001] The invention relates to the field of space intelligent control, in particular to a method for real-time dense reconstruction of space objects. Background technique [0002] With the development of aerospace technology, human exploration and development of outer space resources are getting deeper and deeper. Spacecraft will float freely in space if they fail, fail or are abandoned after completing their missions, that is, they become space junk. Therefore, the space target acquisition technology for the purpose of on-orbit maintenance of traditional spacecraft, space target cleaning, space debris cleaning, space attack and defense, etc. has become a new development direction in the field of space robotics. Obtaining the accurate position information of the target is the premise of realizing its detection, approach, rendezvous and docking and maintenance operations. Therefore, it is particularly important to use the space target 3D reconstruction tech...

AI Technical Summary

Problems solved by technology

However, the traditional real-time dense 3D reconstruction method based on RGB-D images has a better reconstruction effect in a small-scale scene, but it faces a large-scale space scene. The richness of geometric features, the inability to achieve closed-loop optimization, etc.

Images