On the Minimum Attention and the Anytime Attention Control Problems for Linear Systems: A Linear Programming Approach

M. C. F. Donkers,P. Tabuada,W. P. M. H. Heemels,M.C.F. Donkers,W.P.M.H. Heemels

DOI: https://doi.org/10.48550/arXiv.1108.2783

2011-08-13

Optimization and Control

Abstract:In this paper, we present two control laws that are tailored for control applications in which computational and/or communication resources are scarce. Namely, we consider minimum attention control, where the `attention' that a control task requires is minimised given certain performance requirements, and anytime attention control, where the performance under the `attention' given by a scheduler is maximised. Here, we interpret `attention' as the inverse of the time elapsed between two consecutive executions of a control task. By focussing on linear plants, by allowing for only a finite number of possible intervals between two subsequent executions of the control task, by making a novel extension to the notion of control Lyapunov functions and taking these novel extended control Lyapunov function to be infinity-norm-based, we can formulate the aforementioned control problems as online linear programs, which can be solved efficiently. Furthermore, we provide techniques to construct suitable infinity-norm-based extended control Lyapunov functions for our purposes. Finally, we illustrate the resulting control laws using numerical examples. In particular, we show that minimum attention control outperforms an alternative implementation-aware control law available in the literature.