Real-time Autonomous Visual Navigation System for Helicopters Based on GPU Accelerated Speeded-Up Robust Features

H. Bai, X. Xue

Abstract:	This paper extends our previous work named Doppler radar/FOG-IMU/Baro-altimeter integrated navigation system (DFBINS) for helicopters which has been published on ION GNSS 2011 by proposing a novel framework of vision-aided DFBINS autonomous navigation system for helicopters: a visual global positioning system based on previously known geo-referenced images to enable long-range autonomous navigation with a single monocular camera and no need of artificial landmarks or visual beacons. The proposed method is based on image matching between the current helicopter view from a monocular video camera and a previously known database of geo-referenced images. Key points of current video frame and reference database are extracted by implementing Speeded-Up Robust Features (SURF) algorithm on a NVIDIA Geforce 240M graphics processing unit (GPU) and the key points of reference database can be stored in memory before flying off in order to save computing time. Based on the output of DFBINS, a small search area can be confirmed and the key points extracted from current frame can be matched with those stored key points in the indicated search area by using optimized fast library for approximate nearest neighbors (FLANN-search). Afterwards, the best set of inliers, given a fixed computational budget, will be produced and related outliers can be detected and discarded. Flight test was done on a real helicopter and the results confirmed that the proposed system is real-time and sufficiently robust to scale, rotation, illumination, noise and affine projection, thus it can be an ideal scheme for autonomous navigation system for helicopters and other aerial vehicles like AUVs under GPS blockage conditions. The system architecture, software algorithm, computing time, hardware platform and field test results are all described and analyzed in this paper.
Published in:	Proceedings of the 2012 International Technical Meeting of The Institute of Navigation January 30 - 1, 2012 Marriott Newport Beach Hotel & Spa Newport Beach, CA
Pages:	442 - 451
Cite this article:	Bai, H., Xue, X., "Real-time Autonomous Visual Navigation System for Helicopters Based on GPU Accelerated Speeded-Up Robust Features," Proceedings of the 2012 International Technical Meeting of The Institute of Navigation, Newport Beach, CA, January 2012, pp. 442-451.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In