基于BP网络的机器人感觉运动系统研究
Research on Sensorimotor System of Robot Based on BP Neural Network

作者: 戴丽珍 , 阮晓钢 , 于乃功 :北京工业大学电子信息与控制工程学院;

关键词: 机器人BP神经网络感觉运动自主学习Robot BP Neural Network Sensorimotor System Self-Learning

摘要:
生物的诸多技能是在生物个体的生长发育过程中逐渐形成和发展起来的,能否赋予机器人这样一种特性呢?为此,本文基于BP神经网络为机器人建立起一种类似生物的感觉运动系统,使机器人与外界环境之间交互作用,产生自主负趋光行为。实验结果表明机器人在建立的基于BP神经网络的人工感觉运动系统中,能够通过多次的学习,实现自主负趋光行为,并且产生的行为与由生物所产生的行为一致。

Abstract:
Many skills of the natural life are gradually formed and developed during growth and development of indi- vidual organisms. Can a machine be endowed with such characteristics? With the aim, this paper builds a sensorimotor system based on BP neural network for robot. The robot communicates with the environment, and learns negative pho- totaxis. Experiment indicates that the robot with a sensorimotor system based on BP neural network can realize negative phototaxis through several steps of learning, which is the same as the natural life.

文章引用: 戴丽珍 , 阮晓钢 , 于乃功 (2012) 基于BP网络的机器人感觉运动系统研究。 人工智能与机器人研究, 1, 1-5. doi: 10.12677/AIRR.2012.11001

参考文献

[1] J. Houk, A. Gibson. Sensorimotor processing through the cere- bellum. In: J. King, Ed., New Concepts in Cerebellar Neurobi- ology, New York: A.R. Liss, 1987: 387-416.

[2] S. Schaal. Is imitation learning the route to humanoid robots? Trends in Cognitive Sciences, 1999, 3(6): 233-242.

[3] M. Hersch, F. Guenter, S. Calinon and A. Billard. Dy-namical system modulation for robot learning via kinesthetic demon-stra- tions. IEEE Transactions on Robotics, 2008, 24(6): 1463-1467.

[4] H. Hoffmann. Perception through visuomotor antici-pation in a mobile robot. Neural Networks, 2007, 20(1): 22-33.

[5] S. Ekvall, D. Kragic. Robot learning from demonstration: A task-level planning approach. International Journal of Advanced Robotic Systems, 2008, 5(3): 223-234.

[6] 任红格, 史涛, 张瑞成. 基于操作条件反射机制的感觉运动系统认知模型的建立[J]. 机器人, 2012, 3: 292-298.

[7] D. E. Rumelhart, G. E. Hinton and R. J. Williams. Learn-ing repre- sentations by back-propagating errors. Nature, 1986, 323(6088): 533-536.

[8] M. Asada, E. Uchibe and K. Hosoda. Coop-erative behavior ac- quisition for mobile robots in dynamically chang-ing real worlds via vision-based reinforcement learning and develop-ment. Arti- ficial Intelligence, 1999, 110(2): 275-292.

[9] D. Lasser-son, C. Gabriel and B. Sharrack. Nervous system and special senses. Beijing: Science Press, 2002.

[10] K. Gurney, T. J. Prescott and P. Redgrave. A computational mo- del of action selection in the basal ganglia. Ii. Analysis and si- mulation of behaviour. Biological Cyber-netics, 2001, 84: 411- 423.

[11] K. Doya. Complementary roles of basal ganglia and cerebellum in learning and motor control. Current Opinion in Neurobiology, 2000, 10: 732-739.

[12] J. Peters, P. van der Smagt. Searching a scalable approach to cerebellar based control. Ap-plied Intelligence, 2002, 17(1): 11- 33.

分享
Top