An Adaptive Neuromorphic Chip for Combustion Control

Continuous Time Recurrent Neural Networks (CTRNNs) have previously been proposed as an enabling control technology for mechanical devices. Currently, we are in the advanced stages of designing custom VLSI chips that combine automated learning and analog CTRNNs into unified hardware devices capable of learning control laws for physical systems. The chip’s self-configuring capability is potentially useful for the control of combustion systems. In this paper, we will discuss the underlying technology and examine preliminary simulation experiments in which our device successfully learned to suppress instability in a bench top combustor. The paper will conclude with a discussion of expected future work.

[1]  Goldberg,et al.  Genetic algorithms , 1993, Robust Control Systems with Genetic Algorithms.

[2]  John C. Gallagher,et al.  An analysis of the search performance of a mini-population evolutionary algorithm for a robot-locomotion control problem , 2005, 2005 IEEE Congress on Evolutionary Computation.

[3]  John C. Gallagher,et al.  The once and future analog alternative: evolvable hardware and analog computation , 2003, NASA/DoD Conference on Evolvable Hardware, 2003. Proceedings..

[4]  John C. Gallagher.  Evolution and analysis of mixed mode neural networks for walking: mixed pattern generators , 2001, Proceedings of the 2001 Congress on Evolutionary Computation (IEEE Cat. No.01TH8546).

[5]  J J Hopfield,et al.  Neurons with graded response have collective computational properties like those of two-state neurons. , 1984, Proceedings of the National Academy of Sciences of the United States of America.

[6]  Carver Mead,et al.  Analog VLSI and neural systems , 1989 .

[7]  John C. Gallagher,et al.  CTRNN-EH in Silicon: Challenges in Realizing Configurable CTRNNs in VLSI , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[8]  A. Annaswamy,et al.  A Model-Based Active Control Design for Thermoacoustic Instability , 1998 .

[9]  Garrison W. Greenwood,et al.  Introduction to Evolvable Hardware - A Practical Guide for Designing Self-Adaptive Systems , 2006 .

[10]  Randall D. Beer,et al.  On the Dynamics of Small Continuous-Time Recurrent Neural Networks , 1995, Adapt. Behav..

[11]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[12]  Thomas Bäck,et al.  Evolutionary computation: comments on the history and current state , 1997, IEEE Trans. Evol. Comput..

[13]  John C. Gallagher.  Evolution and analysis of non-autonomous neural networks for walking: reflexive pattern generators , 2001, Proceedings of the 2001 Congress on Evolutionary Computation (IEEE Cat. No.01TH8546).

[14]  David B. Fogel,et al.  System Identification Through Simulated Evolution: A Machine Learning Approach to Modeling , 1991 .

[15]  Sanjay K. Boddhu,et al.  A Reconfigurable Analog Neural Network for Evolvable Hardware Applications: Intrinsic Evolution and Extrinsic Verification , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[16]  Yuichi Nakamura,et al.  Approximation of dynamical systems by continuous time recurrent neural networks , 1993, Neural Networks.

[17]  Anuradha M. Annaswamy,et al.  Model-based analysis and design of active control of thermoacoustic instability , 1997, Proceedings of the 1997 IEEE International Conference on Control Applications.