The Fly Algorithm for Robot Navigation
The fly algorithm is a strategy employed for 3D reconstruction of scenes, which employs the genetic algorithms and stereovision principles to determine clusters of points corresponding to different objects present in scene. The obtained reconstruction is partial, but enough to recognize obstacles in the robot space work. This 3D reconstruction strategy also allows to know the dimensions of the detected objects. Many parameters are involved in the fly algorithm, and then it is difficult to assign the optimal values for the best performance. In this work we test different parameters values, analyze the results and present some improvements to the algorithm considering the fly algorithm can be employed in robot navigation.
KeywordsThe fly algorithm Evolutive Strategies Vision stereo Robotics
Unable to display preview. Download preview PDF.
- 2.Gutierrez, S., Marroquin, J.L.: Disparity estimation and reconstruction in stereo vision. Technical communication No I-03-07/7-04-2003. CC/CIMAT, Mexico (2003)Google Scholar
- 3.Quam, L., Hannah, M.J.: Stanford automated photogrammetry research. Technical Report AIM-254, Stanford AI Lab (1974)Google Scholar
- 5.Abdel-Aziz, Y.I., Karara, H.M.: Direct linear transformation from comparator coordinates into object space coordinates in close-range photogrammetry. In: Proc. of the Symp. on Close-Range Photogrammetry, Falls Church, VA, pp. 1–18 (1971)Google Scholar
- 7.Alander, J.T.: Indexed bibliography of genetic algorithms in optics and image processing. Rep. 94-1-OPTICS, Univ. of Vaasa, Dep. of Inf. Tech. and Prod. Ec. (1995)Google Scholar
- 8.Lutton, E., Martinez, P.: A genetic algorithm for the detection of d geometric primitives in images. In: Proceedings of the International Conference on Pattern Recognition, ICPR 1994, Los Alamitos, CA, pp. 526–528 (1994)Google Scholar
- 11.Collet, P., Lutton, E., Raynal, F., Schoenauer, M.: Individual GP: an alternative viewpoint for the resolution of complex problems. In: Proc. of the Genetic and Evolutionary Computation Conference GECCO 1999, San Francisco, CA, vol. 2, pp. 978–981 (1999)Google Scholar