UAV Aerial Mapping with Reduced Frame Overlap and Fast Pose Estimation

The patent segments the mapping process into distinct phases: capturing phase with moderate overlap, initial processing phase to identify stable features, and refinement phase using only those stable features. This segmentation allows the system to achieve high mapping accuracy without processing all overlapping frames, thereby reducing computational demands while maintaining precision.

The patent performs preliminary action by identifying and selecting stable features from captured frames before the main mapping computation. By pre-processing the frame overlap data to extract only reliable features, the system reduces the computational burden during the actual mapping process, thereby decreasing processing time while preserving mapping accuracy.

The patent changes the parameter of frame overlap from the traditional high overlap requirement to a reduced overlap threshold, compensated by enhancing feature selection criteria. By adjusting these parameters and using robust stable feature identification, the system achieves accurate mapping with reduced computational processing demands.

The patent extracts only the essential stable features from the captured frames, separating them from the redundant overlapping data. By taking out only the necessary feature information needed for robust mapping and discarding unnecessary overlap data, the system achieves both mapping robustness and real-time processing capability.