A Method for Registration of 3-D Shapes
IEEE Transactions on Pattern Analysis and Machine Intelligence - Special issue on interpretation of 3-D scenes—part II
Model Based Pose Estimation for Autonomous Operations in Space
ICIIS '99 Proceedings of the 1999 International Conference on Information Intelligence and Systems
Intelligent LIDAR scanning region selection for satellite pose estimation
Computer Vision and Image Understanding
CSCA-based expectivity indices for LIDAR computer vision
MMACTEE'09 Proceedings of the 11th WSEAS international conference on Mathematical methods and computational techniques in electrical engineering
Shape-based pose estimation evaluation using expectivity index artifacts
Proceedings of the Workshop on Performance Metrics for Intelligent Systems
| Hi-index | 0.00 |
This paper presents an innovative approach for the selection of wellconstrained views and surface regions for efficient ICP pose estimation using LIDAR range scanning. The region selection is performed using the Principal Component Analysis technique with derived predictive indices that can be used to assess a view/region for pose estimation. Localized scanning has been proposed for spacecraft rendezvous operations, particularly in the "last mile" scenario where whole object scanning is not possible. The paper illustrates the PCA approach for selection of optimal scanning views and localized regions using (a) CAD models of several spacecraft structures with supporting simulation results based on large amount of data, and (b) a model of a faceted shape, cuboctahedron, which was scanned using Neptec's TriDAR laser scanner. The results confirm the hypothesis that the selected views or regions deliver accurate estimates for the pose norm and also for each component of the pose.