Spatial Registration Based on Weighted Fusion of Multiple Significant Targets

Spatial registration algorithm deeply relies on the significant target. During the whole process of spatial registration, most registration algorithm only select one significant target and the sensor bias remains constant. However, in actual scene, the single significant target cannot be stably detected by all sensors, and the sensor bias will drift over time. Therefore, in order to accurately estimate the sensor bias in real-time, this paper proposes a registration method based on the weighted fusion of multiple significant targets. By fusing the spatial registration results of multiple significant targets, the robustness and accuracy of the registration algorithm are enhanced. Simulation results verify the effectiveness of the proposed algorithm.