Title :
Evolutionary algorithm based neural network controller optimization for autonomous mobile robot navigation
Author :
Han, Seong-Joo ; Oh, Se-young
Author_Institution :
Dept. of Electr. Eng., Pohang Univ. of Sci. & Technol., South Korea
Abstract :
A neural network based navigation algorithm is proposed for mobile robots using ultrasonic sensors. The neural network has a dynamically reconfigurable structure which can not only optimize the weights but also the input sensory connectivity in order to meet any user-defined objective. Further, in order to enhance generalization capability of a single network, a modular network is used in which each network module is optimized for a specific local environment based on environment classification. Both computer simulation and real experiments show the effective performance of the algorithm
Keywords :
digital simulation; evolutionary computation; mobile robots; navigation; neurocontrollers; optimisation; path planning; autonomous mobile robot navigation; computer simulation; dynamically reconfigurable structure; environment classification; evolutionary algorithm based neural network controller optimization; generalization; input sensory connectivity; local environment; modular network; ultrasonic sensors; user-defined objective; Biological cells; Computer simulation; Design optimization; Encoding; Evolutionary computation; Humans; Mobile robots; Navigation; Neural networks; Path planning;
Conference_Titel :
Evolutionary Computation, 2001. Proceedings of the 2001 Congress on
Conference_Location :
Seoul
Print_ISBN :
0-7803-6657-3
DOI :
10.1109/CEC.2001.934380
