Abstract
This paper describes the application of the Structured Genetic Algorithm (sGA) to design neuro-controllers for an unstable physical system. In particular, the approach uses a single unified genetic process to automatically evolve complete neural nets (both architectures and their weights) for controlling a simulated pole-cart system. Experimental results demonstrate the effectiveness of the sGA-evolved neuro-controllers for the task—to keep the pole upright (within a specified vertical angle) and the cart within the limits of the given track.
Similar content being viewed by others
References
C.W. Anderson, “Strategy learning with multilayer connectionist representations,” in Proc. of the Fourth Int. Workshop on Machine Learning, Morgan Kaufmann: Los Altos, 1987, pp. 103-114.
J.E. Baker, “Reducing bias and inefficiency in the selection algorithms,” in Proc. of an Int. Conf. on Genetic Algorithms and Their Applications, Carnegie-Mellon University, Pittsburgh, 1987, pp. 14-21.
A.G. Barto, R.S. Sutton, and C.W. Anderson, “Neuronlike adaptive elements that can solve difficult learning control problems,” IEEE Transactions on Systems, Man and Cybernetics, vol. 13, no.5, pp. 834-846, Sept./Oct. 1983.
H.R. Berenji and P. Khedkar, “Learning and tuning fuzzy logic controllers through reinforcements,” IEEE Transaction on Neural Networks, vol. 3, no.5, pp. 724-740, Sept. 1992.
D. Dasgupta and D.R. McGregor, “Designing application-specific neural networks using the structured genetic algorithm,” in Proc. of the Int. Workshop on Combination of Genetic Algorithms and Neural Networks (COGANN-92), IEEE Computer Society Press: U.S.A, June 1992, pp. 87-96.
D. Dasgupta and D.R. McGregor, “Designing neural networks using the structured genetic algorithm,” in Proc. of the Int. Conf. on Artificial Neural Networks (ICANN), Brighton, U.K., Sept. 1992, pp. 263-268.
D. Dasgupta and D.R. McGregor, “Evolving neurocontrollers for pole balancing,” in Proc. of the Int. Conf. on Artificial Neural Networks (ICANN), Amesterdam, The Netherland, Sept. 1993, pp. 834-837.
D. Dasgupta and D.R. McGregor, “Genetically designing neurocontrollers for a dynamic system,” in Proc. of the Int. Joint Conf. on Neural Networks (IJCNN), Nagoya, Japan, Oct. 1993, pp. 2951-2955.
D. Dasgupta and D.R. McGregor, “A more biologically motivated genetic algorithm: The model and some results,” in Cybernatics and Systems: An International Journal, vol. 25, no.3, pp. 447-469, May-June 1994.
E. Grant and B. Zhang, “A neural-net approach to supervised learning of pole balancing,” in Proc. of IEEE Int. Symposium on Intelligent Control, Albany, New York, Sept. 1989, pp. 123-129.
F.C. Gruau, “Cellular encoding of genetic neural networks,” Technical Report 92-21, Institut IMAG, Grenoble, France, May 1992.
N.J. Hallman, N. Woodcock, and P.D. Picton, “Fuzzy boxes as an alternative to neural networks for difficult problems,” in Application of Artificial Intelligence in Engineering VI (AIENG/91), edited by G. Rzevski and R.A. Adey, pp. 903-919, 1991.
S.A. Harp, T. Samad, and A. Guha, “Designing application-specific neural networks using the genetic algorithm,” Advances in Neural Information Processing Systems, vol. 2, pp. 447-454, 1990.
N. Karunanithi, D. Whitley, and R. Das, “Genetic cascade learning for neural networks,” in Proc. of Int. Workshop on Combinations of Genetic Algorithms and Neural Networks, IEEE Computer Society Press, 1992, pp. 134-145.
H. Kitano, “Designing neural networks using genetic algorithms with graph generation system,” Complex Systems, vol. 4, pp. 461-476, 1990.
H. Kitano, “Neurogenetic learning: An integrated method of designing and training neural networks using genetic algorithms,” Physica D, vol. 75, pp. 225-238, 1994.
J.R. Koza and J.P. Rice, “Genetic generation of both the weights and architecture for a neural network,” in Int. Joint Conf. on Neural Network (IJCNN), 1991.
A. Makarovic, “A Qualitative Way of Solving the Pole Balancing Problem, Oxford University Press, vol. 12, ch. 16, pp. 241-258, 1988.
D.R. McGregor, M.O. Odeytayo, and D. Dasgupta, “Adaptive control of a dynamic system using genetic-based methods,” in IEEE Int. Symposium on Intelligent Control, Glasgow, UK, Aug. 1992, pp. 521-526.
D. Miche and R.A. Chambers, “Boxes: An experiment in adaptive control,” Machine Intelligence, vol. 2, pp. 137-152, 1968.
G.F. Miller, P.M. Todd, and S.U. Hegde, “Designing neural networks using genetic algorithms,” in Proc. of Third Int. Conf. on Genetic Algorithms, edited by J. David Schaffer, pp. 379-384, June 1989.
D.J. Montana and L. Davis, “Training feed forward networks using genetic algorithms,” in Proc. of Int. Joint Conf. on Artificial Intelligence, 1989.
M.O. Odetayo and D.R. McGregor, “Genetic algorithm for control rules for a dynamic system,” in Proc. of the Int. Conf. on Genetic Algorithms (ICGA), 1989, pp. 177-182.
I.C. Parmee, “Diverse evolutionary search for preliminary whole-system design,” in Proc. of the 4th Int. Conf. on AI in Civil and Structural Engineering, 1995.
C. Sammut, “Experimental results from an evaluation of algorithms that learn to control dynamic systems,” in Proc. of the Fifth Int. Conf. on Machine Learning, 1988.
D. Thierens and L. Vercauteren, “A topology exploiting genetic algorithm to control dynamic systems,” in Lecture Notes in Computer Science, edited by G. Goos and Hartmanis, Springer-Verlag, pp. 104-108, 1991, Proc. of PPSN-1, 1990.
D. Whitley and T. Starkweather, “Optimizing small neural networks using a disributed genetic algorithm,” in Int. Joint Conf. on Neural Network (IJCNN), 1990, pp. 206-209.
D. Whitley, T. Starkweather, and C. Bogart, “Genetic algorithms and neural networks: Optimizating connections and connectivity,” Parallel Computing, vol. 14, pp. 347-361, 1990.
D. Whitley, S. Dominic, and R. Das, “Genetic reinforcement learning with multilayer neural networks,” in 4th Int. Conf. on Genetic Algorithms, 1991, pp. 562-569.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Dasgupta, D. Evolving Neuro-Controllers for a Dynamic System Using Structured Genetic Algorithms. Applied Intelligence 8, 113–121 (1998). https://doi.org/10.1023/A:1008291923124
Issue Date:
DOI: https://doi.org/10.1023/A:1008291923124