Title :
Immunized neurocontrol-concepts and initial results
Author :
KrishnaKumar, K.
Author_Institution :
Dept. of Aerosp. Eng., Alabama Univ., Tuscaloosa, AL, USA
fDate :
6/6/1992 12:00:00 AM
Abstract :
The role of artificial neural networks in control of complex systems is seeing a rapid growth due to the potential of these networks to emulate complex, non-linear systems and thus help in the automated control of such systems. The paper addresses an important area in automatic control, namely, adaptive control. Strong connections between concepts from immunology, genetic algorithms, and adaptive neurocontrol are drawn. Immunology is the science of in-built defense mechanism that is present in all living beings to protect them against external attacks. The science of immunology has many parallels to the robust adaptive control problem. Some of these parallels are presented and based on these parallels, a procedure to realize an immunized neurocontrol structure is developed. Initial results of the implementation of this procedure in adaptive control of an uncertain UH-1 longitudinal helicopter model are included
Keywords :
adaptive control; genetic algorithms; neural nets; adaptive control; artificial neural networks; automatic control; complex systems; genetic algorithms; immunized neurocontrol; uncertain UH-1 longitudinal helicopter model; Adaptive control; Adaptive systems; Artificial neural networks; Automatic control; Control systems; Evolution (biology); Genetic algorithms; Immune system; Nonlinear control systems; Robust control;
Conference_Titel :
Combinations of Genetic Algorithms and Neural Networks, 1992., COGANN-92. International Workshop on
Conference_Location :
Baltimore, MD
Print_ISBN :
0-8186-2787-5
DOI :
10.1109/COGANN.1992.273941
