An Adaptive Neuromorphic Chip for Combustion Control
John Gallagher, Mitch Wolff · 2007
Continuous Time Recurrent Neural Networks (CTRNNs) have previously been proposedas an enabling control technology for mechanical devices. Currently, we are in the advancedstages of designing custom VLSI chips that combine automated learning and analogCTRNNs into unified hardware devices capable of learning control laws for physicalsystems. The chip’s self-configuring capability is potentially useful for the control ofcombustion systems. In this paper, we will discuss the underlying technology and examinepreliminary simulation experiments in which our device successfully learned to suppressinstability in a bench top combustor. The paper will conclude with a discussion of expectedfuture work.