Controlling Chaos Using Edge Computing Hardware
This work addresses the need for efficient, low-power machine learning controllers for autonomous systems at the edge, representing an incremental step in deploying such algorithms.
The paper tackled the problem of controlling chaotic systems to arbitrary time-dependent states using a nonlinear controller based on next-generation reservoir computing, achieving a model that requires only 25.0 ± 7.0 nJ per evaluation and is deployable on embedded hardware like FPGAs.
Machine learning provides a data-driven approach for creating a digital twin of a system - a digital model used to predict the system behavior. Having an accurate digital twin can drive many applications, such as controlling autonomous systems. Often the size, weight, and power consumption of the digital twin or related controller must be minimized, ideally realized on embedded computing hardware that can operate without a cloud-computing connection. Here, we show that a nonlinear controller based on next-generation reservoir computing can tackle a difficult control problem: controlling a chaotic system to an arbitrary time-dependent state. The model is accurate, yet it is small enough to be evaluated on a field-programmable gate array typically found in embedded devices. Furthermore, the model only requires 25.0 $\pm$ 7.0 nJ per evaluation, well below other algorithms, even without systematic power optimization. Our work represents the first step in deploying efficient machine learning algorithms to the computing "edge."