At this point in our research, we felt that the successful development of a cartpole fuzzy controller was a nice milestone. After all, we had developed a system that could use linguistic rules similar to those used by human operators to manipulate a system that is difficult to control. However, we also felt that we had a number of questions that needed exploring. Some of the questions were fairly obvious like, can this approach be used on a real problem? Are there better ways to select the rules; the membership functions? Some of the questions were not quite as obvious and required more subtle answers such as what happens when the system has additional parameters that change periodically such as the mass of the cart and the length of the pole? In the remainder of this book we will address these questions and others. However, it will take us a little time to get to them all. In the next chapter we will discuss an opportunity we were presented with at about this time in the project; we will discuss the development of a controller for a space satellite rendezvous system.

References

Larkin, L. I. (1985). A fuzzy logic controller for aircraft flight control. Industrial Applications of Fuzzy Control (M. Sugeno, ed.), Amsterdam: North-Holland, p. 87.

Karr, C. L., Fleming, J. W., & Vann, P. A. (1994). Aspects of genetic algorithm-designed fuzzy logic controllers (Report of Investigations number 9515). Washington, DC: U.S. Department of the Interior, Bureau of Mines.

Press, W. H., Flannery, B. P., Teukolsky, S. A., and Vetterling, W. T. (1988). Numerical recipes in C. Cambridge, MA: Cambridge University Press, p. 567.