This paper outlines a method for the automatic creation of control systems and a search to find the optimal controller through the optimisation of candidate solutions. A software system to carry out Parallel and Distributed Evolutionary Computations (PDEC) is introduced. This system is designed to interface with Simulink which is used as the simulation environment. It has the potential to create a control system made up of any number of blocks from a Simulink library including specialised blocks. The paper then describes how the software carries out the search and low this can be run in parallel. It also introduces current work involving real-time search with hardware-in-the-loop. Finally a speed controller for an Electric Thrust Reversal Actuation System is designed using the system and tested showing the results.
[1]
Marc Parizeau,et al.
Distributed Beagle: An Environment For Parallel And Distributed Evolutionary Computations
,
2003
.
[2]
Piero P. Bonissone,et al.
Genetic algorithms for automated tuning of fuzzy controllers: a transportation application
,
1996,
Proceedings of IEEE 5th International Fuzzy Systems.
[3]
Peter J. Fleming,et al.
The MATLAB genetic algorithm toolbox
,
1995
.
[4]
Fred Glover,et al.
Tabu Search and Adaptive Memory Programming — Advances, Applications and Challenges
,
1997
.
[5]
John R. Koza,et al.
Automatic Creation of Human-Competitive Programs and Controllers by Means of Genetic Programming
,
2000,
Genetic Programming and Evolvable Machines.