Robust almost sure stability for uncertain stochastically scheduled anytime controllers

In this paper we consider closed loop stability of a number of different software tasks implementing a hierarchy of real-time controllers for a given plant, i.e. "anytime controllers". The execution of the control tasks are driven by the available computational time of an embedded platform under stringent real-time constraints. Hence, preemptive scheduling schemes are considered, under which the maximum execution time allowed for control software tasks is not a-priori known. A stochastic description of the scheduler accounting for the presence of some uncertainties in the model, is provided. An anytime control hierarchy of controllers for the same plant, in which higher controllers in the hierarchy provide better closed-loop performance but require larger worst-case execution times, is assumed to be given. Since the ensuing switching system is prone to instability, the presented method allows to robustly condition the partially known stochastic scheduler so as to obtain a better exploitation of computing capabilities, while guaranteeing almost sure stability of the resulting switching system.

[1]  Bonnie H. Ferri,et al.  Digital Filters with Adaptive Length for Real-Time Applications , 2004 .

[2]  Robert K. Brayton,et al.  Stability of dynamical systems: A constructive approach , 1979 .

[3]  G. Nicolao,et al.  On almost sure stability of discrete-time Markov jump linear systems , 2004, 2004 43rd IEEE Conference on Decision and Control (CDC) (IEEE Cat. No.04CH37601).

[4]  R. Decarlo,et al.  Solution of coupled Lyapunov equations for the stabilization of multimodal linear systems , 1997, Proceedings of the 1997 American Control Conference (Cat. No.97CH36041).

[5]  Karl-Erik Årzén,et al.  The Jitter Margin and Its Application in the Design of Real-Time Control Systems , 2004 .

[6]  A. Morse,et al.  Stability of switched systems: a Lie-algebraic condition ( , 1999 .

[7]  K Furuta,et al.  Swing-up Control of Inverted Pendulum Using Pseudo-State Feedback , 1992 .

[8]  João Pedro Hespanha,et al.  Switching between stabilizing controllers , 2002, Autom..

[9]  A. Michel,et al.  Stability theory for hybrid dynamical systems , 1995, Proceedings of 1995 34th IEEE Conference on Decision and Control.

[10]  K. Loparo,et al.  Almost sure and δmoment stability of jump linear systems , 1994 .

[11]  S. Pettersson,et al.  Stabilization of hybrid systems using a min-projection strategy , 2001, Proceedings of the 2001 American Control Conference. (Cat. No.01CH37148).

[12]  Giorgio Buttazzo,et al.  Hard Real-Time Computing Systems: Predictable Scheduling Algorithms and Applications , 1997 .

[13]  P. Antsaklis,et al.  Design of stabilizing switching control laws for discrete- and continuous-time linear systems using piecewise-linear Lyapunov functions , 2002 .

[14]  Riccardo Bettati,et al.  Imprecise computations , 1994, Proc. IEEE.

[15]  James W. Layland,et al.  Scheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment , 1989, JACM.

[16]  Antonio Bicchi,et al.  Anytime Control Algorithms for Embedded Real-Time Systems , 2008, HSCC.

[17]  Robert K. Brayton,et al.  Constructive stability and asymptotic stability of dynamical systems , 1980 .

[18]  Antonio Bicchi,et al.  Almost sure stability of anytime controllers via stochastic scheduling , 2007, 2007 46th IEEE Conference on Decision and Control.

[19]  R. Decarlo,et al.  Perspectives and results on the stability and stabilizability of hybrid systems , 2000, Proceedings of the IEEE.

[20]  M. Branicky.  Stability of hybrid systems: state of the art , 1997, Proceedings of the 36th IEEE Conference on Decision and Control.