Abstract An improved warping algorithm is proposed for visual robot homing. The original warping algorithm has to search the whole parameter space, which results in the problem of high computational cost and limited application scope. To solve this problem, a gradient descent method based on randomly selected initial points is implemented, which significantly decreases the tested parameters and improves the algorithm efficiency. Experimental results of several real scenes show that the proposed algorithm significantly improves the computational efficiency by one order of magnitude with the robustness.
[1] Cartwright B A, Collett T S. Landmark Learning in Bees. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology, 1983, 151(4): 521-543 [2] Davison A J. Real-Time Simultaneous Localization and Mapping with a Single Camera // Proc of the 9th IEEE International Conference on Computer Vision. Nice, France, 2003, Ⅱ: 1403-1410 [3] Eade E, Drummond T. Scalable Monocular Slam // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, USA, 2006, Ⅰ: 469-476
[4] Vardy A. Long-Range Visual Homing // Proc of the IEEE International Conference on Robotics and Biomimetics. Kunming, China, 2006: 220-226 [5] Goedeme T, Nuttin M, Tuytelaars T. Omnidirectional Vision Based Topological Navigation. International Journal of Computer Vision, 2007, 74(3): 219-236 [6] Cassinis R, Duina D, Inelli S, et al. Unsupervised Matching of Visual Landmarks for Robotic Homing Using Fourier-Mellin Transform. Robotics and Autonomous Systems, 2002, 40(2/3): 131-138 [7] Pons J S, Hubner W, Dahmen J. Vision-Based Robot Homing in Dynamic Environments // Proc of the 13th IASTED International Conference on Robotics and Applications. Wuerzburg, Germany, 2007: 293-298 [8]Argyros A, Bekris K, Orphanoudakis S. Robot Homing Based on Corner Tracking in a Sequence of Panoramic Images // Proc of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Hawaii, USA, 2001: 3-10 [9] Vardy A, Mller R. Biologically Plausible Methods for Robot Visual Homing. Connection Science , 2005, 17(1): 47-89 [10] Zeil J, Hofmann M, Chahl J. Catchments Areas of Panoramic Snapshots in Outdoor Scenes. The Optical Society of America, 2003, 20(3): 450-469 [11] Stürzl W, Zeil J. Depth, Contrast and View-Based Homing in Outdoor Scenes. Biological Cybernetics, 2007, 96(5): 519-531 [12] Mller R, Vardy A. Local Visual Homing by Matched-Filter Descent in Image Distances. Biological Cybernetics, 2006, 95(5): 413-430 [13]Binding D, Labrosse F. Visual Local Navigation Using Warped Panoramic Images // Proc of the Conference Towards Autonomous Robotic Systems. Guildford, UK, 2006: 19-26 [14] Franz M O, Schlkopf B, Mallot H A. Where Did I Take That Snapshot? Scene-Based Homing by Image Matching. Biological Cybernetics, 1998, 79(3): 191-202 [15] Sturzl W, Mallot H A. Efficient Visual Homing Based on Fourier Transformed Panoramic Images. Robotics and Autonomous Systems, 2006, 54(4): 300-313 [16] Labrosse F. Visual Compass // Proc of the Conference Towards Autonomous Robotic Systems. Colchester, UK, 2004: 85-92 [17] Mitchell T, Labrosse F. Visual Homing: A Purely Appearance-Based Approach // Proc of the Conference Towards Autonomous Robotic Systems. Colchester, UK, 2004: 101-108 [18] Mller R. Local Visual Homing by Warping of Two-Dimensional Images. Robotics and Autonomous Systems, 2009, 57(1): 87-101 [19] Churchill D, Vardy A. Homing in Scale Space // Proc of the IEEE/RSJ International Conference on Intelligent Robots and Systems. Nice, France, 2008: 1307-1312
2010, Vol. 23 
