Particle Filter Approach to Vision-Based Navigation with Aerial Image Segmentation
Abstract
This study proposes a novel approach for a vision-based navigation problem using semantically segmented aerial images generated by a convolutional neural network. Vision-based navigation provides a position solution by matching an aerial image to a georeferenced database, and it has been increasingly studied for global navigation satellite system–denied environments. Aerial images include a vast amount of information that infers the position where they are located. However, it also includes features that disturb the estimation accuracy. The progress of convolutional neural network may provide a promising solution for extracting only helpful features for this purpose. Therefore, segmented images are modeled as a Gaussian mixture model, and the distance for a quantitative discrepancy between two images is established. This allows us to compare the two images quickly with improved accuracy. In addition, a framework of a particle filter is applied to estimate the position using an inertial navigation system. It employs the distance as a measurement, and the particles tend to converge to the true position. Flight test experiments were conducted to verify that the proposed approach achieved distance error of less than 10 m.
References
[1] , “Vision-Based UAVs Aerial Image Localization: A Survey,” Proceedings of the 2nd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery, Assoc. for Computing Machinery, New York, Nov. 2018, pp. 9–18. https://doi.org/10.1145/3281548.3281556
[2] , “Vision-Based Unmanned Aerial Vehicle Navigation Using Geo-Referenced Information,” EURASIP Journal on Advances in Signal Processing, Vol. 2009, Jan. 2009, pp. 1–18. https://doi.org/10.1155/2009/387308
[3] , “A New Paradigm for Matching UAV and Aerial Images,” ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. 3, July 2016, pp. 83–90. https://doi.org/10.5194/isprs-annals-III-3-83-2016
[4] , “Fully Convolutional Networks for Semantic Segmentation,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE, New York, 2015, pp. 3431–3440.
[5] , “Mask R-CNN,” Proceedings of the IEEE International Conference on Computer Vision, IEEE, New York, 2017, pp. 2961–2969.
[6] , “Cross-View Image Matching for Geo-Localization in Urban Environments,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE, New York, 2017, pp. 3608–3616.
[7] , “Vision-Based Navigation Using Gaussian Mixture Model of Terrain Features,” AIAA Scitech 2020 Forum, AIAA Paper 2020-1344, Jan. 2020. https://doi.org/10.2514/6.2020-1344
[8] , “A Deep CNN-Based Framework for Enhanced Aerial Imagery Registration with Applications to UAV Geolocalization,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, IEEE, New York, 2018, pp. 1513–1523.
[9] , “GPS-Denied UAV Localization Using Pre-Existing Satellite Imagery,” 2019 International Conference on Robotics and Automation (ICRA), IEEE, New York, May 2019, pp. 2974–2980.
[10] , “BRM Localization: UAV Localization in GNSS-Denied Environments Based on Matching of Numerical Map and UAV Images,” arXiv preprint arXiv:2008.01347.
[11] , “A Tutorial on Particle Filters for Online Nonlinear/Non-Gaussian Bayesian Tracking,” IEEE Transactions on Signal Processing, Vol. 50, No. 2, Aug. 2002, pp. 174–188. https://doi.org/10.1109/78.978374
[12] , “Ordinal Measures for Image Correspondence,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 20, No. 4, April 1998, pp. 415–423. https://doi.org/10.1109/34.677275
[13] , “Distinctive Image Features from Scale-Invariant Keypoints,” International Journal of Computer Vision, Vol. 60, No. 2, Nov. 2004, pp. 91–110. https://doi.org/10.1023/B:VISI.0000029664.99615.94
[14] , “Speeded-Up Robust Features (SURF),” Computer Vision and Image Understanding, Vol. 110, No. 3, June 2008, pp. 346–359. https://doi.org/10.1016/j.cviu.2007.09.014
[15] , “ORB: An Efficient Alternative to SIFT or SURF,” 2011 International Conference on Computer Vision, IEEE, New York, Nov. 2011, pp. 2564–2571. https://doi.org/10.1109/ICCV.2011.6126544
[16] , “Comparison of OpenCV’s Feature Detectors and Feature Matchers,” 2016 23rd International Conference on Mechatronics and Machine Vision in Practice (M2VIP), IEEE, New York, Nov. 2016, pp. 1–6. https://doi.org/10.1109/M2VIP.2016.7827292
[17] , “The Multivariate Gaussian Probability Distribution,” Tech. Rept., Technical Univ. of Denmark, 2005.
[18] , Applied Mathematics in Integrated Navigation Systems, AIAA, Reston, VA, 2007, Chap. 14. https://doi.org/10.2514/4.861598
[19] , “A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise,” KDD 1996 Proceedings of the Second International Conference on Knowledge Discovery and Data Mining, KDD-96, Vol. 96, AAAI Press, Menlo Park, CA, 1996, pp. 226–231.
[20] , “Comparison of Resampling Schemes for Particle Filtering,” ISPA 2005. Proceedings of the 4th International Symposium on Image and Signal Processing and Analysis, IEEE, New York, Sept. 2005, pp. 64–69. https://doi.org/10.1109/ISPA.2005.195385