DocumentCode :
2934012
Title :
The binary-weights neural network for robot control
Author :
Li, Shuguang ; Yuan, Jianping ; Yue, Xiaokui ; Luo, Jianjun
Author_Institution :
Northwestern Polytech. Univ., Xi´´an, China
fYear :
2010
fDate :
26-29 Sept. 2010
Firstpage :
765
Lastpage :
770
Abstract :
We propose a pure topological recurrent networks controller, which has random binary connections in hidden layer, and all hidden neurons are activated by sinusoidal functions. A direct graph encoding method and four genetic operators are implemented for using genetic programming to train this controller. Firstly, its feasibility and efficiency were validated by a pair of function approximation experiments, the results show that through evolutionary learning, this novel RNN controller can handle nonlinear problems as well as common RNN even without adjustable weights. Moreover, a simulated mobile robot was equipped with this controller, and the robot was navigated around obstacles toward a goal in physical simulation environments; during tests, this robot exhibited four successful behaviors just by topological evolving on the simple controller. This experiment reveals that this controller has the simplicity, usability and potential for robot control, it then raises the hope for further works in exploring network motifs from high level controllers.
Keywords :
collision avoidance; encoding; genetic algorithms; learning (artificial intelligence); mobile robots; neurocontrollers; random functions; recurrent neural nets; binary-weight neural network; direct graph encoding; evolutionary learning; genetic operator; genetic programming; navigation; nonlinear problems; obstacle avoidance; random binary connection; robot control; simulated mobile robot; sinusoidal functions; topological recurrent artificial neural network controller; Artificial neural networks; Mobile robots; Neurons; Recurrent neural networks; Robot kinematics; Training;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Biomedical Robotics and Biomechatronics (BioRob), 2010 3rd IEEE RAS and EMBS International Conference on
Conference_Location :
Tokyo
ISSN :
2155-1774
Print_ISBN :
978-1-4244-7708-1
Type :
conf
DOI :
10.1109/BIOROB.2010.5626893
Filename :
5626893
Link To Document :
بازگشت