Multiagent fuzzy-neural control of a 3-link uniped

Kgroo, a simulated 3-link folding legged uniped robot is presented and locomotion training of Kgroo with fuzzy-neural control is discussed. It is observed that for the uniped locomotion problem, global training of a fuzzy or neural controller is subject to failure. It is shown that, starting with a single jump example, a multiagent cerebellum model (MAC-J) can enable Kgroo to learn different jumps with a geometrical learning rate based on a learning-tuning-brainstorming theory. Technically, this work introduces effective means for decomposing the high-dimensional locomotion control problem into kernel spaces; theoretically, incremental learning and coordinated cerebellar agent discovery provide a natural explanation to certain explosive learning behaviors in human and animal locomotion control.

[1]  Wen-Ran Zhang,et al.  Nesting, safety, layering, and autonomy: a coordinated computational intelligence (CCI) approach to folding legged robot locomotion and gymnastic training , 1996, Proceedings of the 1996 IEEE International Symposium on Intelligent Control.

[2]  I. J. Schoenberg.  Contributions to the Problem of Approximation of Equidistant Data by Analytic Functions , 1988 .

[3]  J. Buckley,et al.  Fuzzy input-output controllers are universal approximators , 1993 .

[4]  Wen-Ran Zhang,et al.  Locomotion and gymnastic training of a 4-link uniped with coordinated computational intelligence (CCI) , 1996, Proceedings of IEEE 5th International Fuzzy Systems.

[5]  James C. Bezdek,et al.  On the relationship between neural networks, pattern recognition and intelligence , 1992, Int. J. Approx. Reason..

[6]  H. W. Werntges.  Partitions of unity improve neural function approximators , 1993, IEEE International Conference on Neural Networks.

[7]  James C. Bezdek,et al.  Pool2: a generic system for cognitive map development and decision analysis , 1989, IEEE Trans. Syst. Man Cybern..

[8]  Jerry M. Mendel,et al.  Generating fuzzy rules by learning from examples , 1992, IEEE Trans. Syst. Man Cybern..

[9]  Marc H. Raibert,et al.  Legged Robots That Balance , 1986, IEEE Expert.

[10]  James S. Albus,et al.  New Approach to Manipulator Control: The Cerebellar Model Articulation Controller (CMAC)1 , 1975 .

[11]  H. Wang,et al.  A neuromorphic controller for a three-link biped robot , 1989, International 1989 Joint Conference on Neural Networks.

[12]  Edmund H. Durfee,et al.  Trends in Cooperative Distributed Problem Solving , 1989, IEEE Trans. Knowl. Data Eng..

[13]  Wen-Ran Zhang,et al.  NPN fuzzy sets and NPN qualitative algebra: a computational framework for bipolar cognitive modeling and multiagent decision analysis , 1996, IEEE Trans. Syst. Man Cybern. Part B.

[14]  W. T. Miller.  Learning dynamic balance of a biped walking robot , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).

[15]  W. Pedrycz.  Why triangular membership functions , 1994 .

[16]  W.-R. Zhang,et al.  A cognitive-map-based approach to the coordination of distributed cooperative agents , 1992, IEEE Trans. Syst. Man Cybern..

[17]  J. Raczkowsky,et al.  Emulation of spline curves and its applications in robot motion control , 1994, Proceedings of 1994 IEEE 3rd International Fuzzy Systems Conference.

[18]  Liqun Andrew Li.  Uniped locomotion training , 1996 .

[19]  Jianwei Zhang,et al.  Modular design of fuzzy controller integrating deliberative and reactive strategies , 1996, Proceedings of IEEE International Conference on Robotics and Automation.

[20]  Wenhua Wang,et al.  A‐pool: An agent‐oriented open system shell for distributed decision process modeling , 1994 .