JaeHyung Kim1/Joon Lyou/HwyKuen Kwak
The Journal of Mechanical Science and Technology, vol. 24, no. 2, pp.489-496, 2010

Abstract : This paper presents a framework for a GPS/INS/vision-based helicopter navigation system. The conventional GPS/INS algorithm has weak points such as GPS blockage and jamming, while the helicopter is a speedy and highly dynamic vehicle that may easily lose a GPS signal. A vision sensor is not affected by signal jamming, and the navigation error of such a system does not accumulate. Hence, a GPS/INS/vision-aided navigation scheme was implemented to provide the robust localization suitable for helicopter operations in various environments. The core algorithm is the vision-based SLAM (simultaneous localization and mapping) technique. Flight tests were performed to verify the SLAM-aided vision navigation algorithm. During the tests, it was confirmed that the developed system is sufficiently robust under GPS blockage conditions. The system design, software algorithm, and flight test results are described in this paper.

Keyword : GPS/INS/Vision navigation; SLAM; Kalman filter; Harris corner; SIFT