Xin Yao's Research Interests: Evolving Rule-based Systems for Nonlinear Control Problems


Many nonlinear control problems are hard to solve analytically. They are also difficult to solve using conventional optimsation or programming methods. Evolutionary algorithms can play a significant role in tackling such problems. In our own work, we have concentrated on evolving high-level rules that are comprehensible by human beings so that the evolved rules can be analysed and the knowledge transferred to solving other problems.


Selected Papers

  1. J. Bobbin and X. Yao, ``Evolving rules for nonlinaer control,'' In New Frontier in Computational Intelligence and its Applications, Masoud Mohammadian (ed.), IOS Press, Amsterdam, 2000, pp.197-202. (ISBN 90 5199 476 1)
    (Available as a gzipped ps file.)

  2. J. Bobbin and X. Yao, ``Automatic Discovery of Relational Information in Comprehensible Control Rules by Evolutionary Algorithms,'' Proceedings of the Third Australia-Japan Joint Workshop on Intelligent and Evolutionary Systems, Canberra, Australia, 23-25 November 1999, pp.117-123.

  3. J. Bobbin and X. Yao, ``Solving optimal control problems with a cost on changing control by evolutionary algorithms,'' Proc. of 1997 IEEE International Conference on Evolutionary Computation (ICEC'97), 13-16 April 1997, Indianapolis, USA, pp.331--336.
    Available as bobbin_yao_icec97.ps.gz.