8 February 2017 Visual identification and similarity measures used for on-line motion planning of autonomous robots in unknown environments
Author Affiliations +
Proceedings Volume 10225, Eighth International Conference on Graphic and Image Processing (ICGIP 2016); 102251M (2017) https://doi.org/10.1117/12.2266997
Event: Eighth International Conference on Graphic and Image Processing, 2016, Tokyo, Japan
Abstract
In this paper we propose an on-line motion planning strategy for autonomous robots in dynamic and locally observable environments. In this approach, we first visually identify geometric shapes in the environment by filtering images. Then, an ART-2 network is used to establish the similarity between patterns. The proposed algorithm allows that a robot establish its relative location in the environment, and define its navigation path based on images of the environment and its similarity to reference images. This is an efficient and minimalist method that uses the similarity of landmark view patterns to navigate to the desired destination. Laboratory tests on real prototypes demonstrate the performance of the algorithm.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Fredy Martínez, Fredy Martínez, Fernando Martínez, Fernando Martínez, Edwar Jacinto, Edwar Jacinto, } "Visual identification and similarity measures used for on-line motion planning of autonomous robots in unknown environments", Proc. SPIE 10225, Eighth International Conference on Graphic and Image Processing (ICGIP 2016), 102251M (8 February 2017); doi: 10.1117/12.2266997; https://doi.org/10.1117/12.2266997
PROCEEDINGS
5 PAGES


SHARE
Back to Top