RT Conference Proceedings
T1 Applying evolution strategies to neural networks robot controller
A1 Berlanga de Jesús, Antonio
A1 Molina López, José Manuel
A1 Sanchis de Miguel, María Araceli
A1 Isasi, Pedro
AB In this paper an evolution strategy (ES) is introduced, to learn weights of a neural network controller in autonomous robots. An ES is used to learn high-performance reactive behavior for navigation and collisions avoidance. The learned behavior is able to solve the problem in different environments; so, the learning process has proven the ability to obtain a specialized behavior. All the behaviors obtained have been tested in a set of environment and the capability of generalization is showed for each learned behavior. No subjective information about “how to accomplish the task” has been included in the fitness function. A simulator based on mini-robot Khepera has been used to learn each behavior.
PB Springer
SN 978-3-540-66068-2
SN 1611-3349 (Online)
YR 1999
FD 1999
LK https://hdl.handle.net/10016/4006
UL https://hdl.handle.net/10016/4006
LA eng
NO Proceeding of: International Work-Conference on Artificial and Natural Neural Networks, IWANN'99 Alicante, Spain, June 2–4, 1999
DS e-Archivo
RD 19 may. 2024