Fully parallel learning neural network chip for real-time control
Presented in this paper is a parallel learning neural network chip, which is used to perform real-time output feedback control on a nonlinear dynamic plant. The controlled plant is a simulated unstable combustion process. Neural networks provide an adaptive sub-optimal control that does not need any prior knowledge of the system. In addition, the hardware neural network presented here utilizes parallelism to achieve speed independent of the size of the network, enabling real-time control. On-chip learning ability allows the hardware neural network to learn on-line as the plant is running and the plant parameters are changing. Also described is the experimental setup used to obtain the results.