Vision-based autonomous landing integrated navigation scheme of unmanned combat aerial vehicles

A computer-vision-based integrated navigation scheme for autonomous landing of an unmanned combat air vehicle (UCAV) was presented. Based on this scheme, vision information with measurements of other on-board sensors, including inertial navigation system (INS) and altimeter system, could be fused according to the characteristics of these sensor systems. The scheme gave navigation information without the help of external navigation equipments with high measurement accuracy. Since computer vision played an important role in this navigation scheme, the vision algorisms were complicated processes and were discussed, so that the vision sensor measurement could be output with a delay in a low bandwidth. A multi-rate extended Kalman filter was constructed to fuse multi-rate information and gave high bandwidth attitude and pose estimations based on the output bandwidth of INS. The navigation scheme could run properly on the real-time simulation system for autonomous landing of the UCAV.