This paper puts forward using CCD camera to recognize color path information and realize vision navigation, which simplifies the image feature abstraction and improves the robust and reliability of path recognition than gray image recognizing. Additional, in this paper, we design new fuzzy control strategy, when the curving extent of path satisfies definite condition, adopting fuzzy control based on genetic algorithm (GA) optimizing. With GA off-line optimizing the membership function parameters of fuzzy controller, consequently, keeps the simple, vivid and fast characters of fuzzy control and strengthens its self-adaptability. This strategy of road following has already been validates on simulation platform and successfully applied on the HEBUT-I intelligent mobile robot. The result proves its validity and feasibility
[1]
Nasser Kehtarnavaz,et al.
Visual control of an autonomous vehicle (BART)-the vehicle-following problem
,
1991
.
[2]
G. Langholz,et al.
Genetic-Based New Fuzzy Reasoning Models with Application to Fuzzy Control
,
1994,
IEEE Trans. Syst. Man Cybern. Syst..
[3]
Romuald Aufrère,et al.
A Dynamic Vision Algorithm to Locate a Vehicle on a Nonstructured Road
,
2000,
Int. J. Robotics Res..
[4]
Ying Sun,et al.
A hierarchical approach to color image segmentation using homogeneity
,
2000,
IEEE Trans. Image Process..
[5]
Julian F. Y. Cheung,et al.
Directional line detectors in correlated noisy environments
,
2000,
IEEE Trans. Image Process..