INTELLIGENT ADAPTIVE CONTROL OF NON-LINEAR SYSTEMS BASED ON EMOTIONAL LEARNING APPROACH
Ali Reza Mehrabian, Caro Lucas · International Journal of Artificial Intelligence Tools · 2007
Drawing upon the past research on neuromorphic implementation of an emotionally intelligent control system, a developed version of the brain emotional learning based intelligent controller, BELBIC, is proposed in this paper. The modified system, which essentially realizes direct adaptive control scheme, deals very well for SISO plants, and does not need any pre-training. It is adaptive and robust with respect to changes in the environment since it learns to produce appropriate control actions on-line. An additional advantage of BELBIC is its simplicity as well as low computational load. The performance of the system is demonstrated via application on previously studied benchmarks involving optimization and adaptive control in nonlinear systems with MLP-backpropagation (BP) nets.