Resource efficient stochastic predictive control

What is it about?

This article presents a resource efficient framework for a class of stochastic control systems that utilizes state dependent control strategies in order to reduce the online computational load. When the states are in a given neighbourhood of the desired operating point, the controller is switched off, and data from the feedback channel is not transmitted. Outside this neighbourhood, we pay close attention to the performance of the controller by adopting a stochastic predictive algorithm when the states are in a predefined \emph{comfort} zone, and activate a recovery algorithm beyond the comfort zone that secures at least good qualitative properties. We demonstrate that our proposed controller leads to mean square boundedness of the closed loop states in the presence of stochastic noise, bounded control authority, and control channel erasures, while entailing a dramatic reduction in network traffic and computational resources.

Featured Image

Why is it important?

The method proposed in this article gives significant improvement in savings of actuator energy, reduction in transmissions of data through the channel and efficiently utilizes the computational resources.

Perspectives