SUB-OPTIMAL SENSOR SCHEDULING WITH ERROR BOUNDS

Abstract In this paper the problem of sub-optimal sensor scheduling with a guaranteed distance to optimality is considered. Optimal in the sense that the sequence that minimizes the estimation error covariance matrix for a given time horizon is found. The search is done using relaxed dynamic programming. The algorithm is then applied to one simple second order system and one sixth order model of a fixed mounted model lab helicopter.

[1]  M. Athans.  On the Determination of Optimal Costly Measurement Strategies for Linear Stochastic Systems , 1972 .

[2]  Efstratios Skafidas,et al.  Optimal measurement scheduling in linear quadratic Gaussian control problems , 1998, Proceedings of the 1998 IEEE International Conference on Control Applications (Cat. No.98CH36104).

[3]  Bo Lincoln,et al.  Dynamic Programming and Time-Varying Delay Systems , 2003 .

[4]  Joel W. Burdick,et al.  Scheduling for distributed sensor networks with single sensor measurement per time step , 2004, IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004.

[5]  Andrew E. B. Lim,et al.  Sensor scheduling in continuous time , 2001, Autom..

[6]  Magnus Gäfvert.  Modelling of the ETH Helicopter Laboratory Process , 2001 .

[7]  Robin J. Evans,et al.  The problem of optimal robust sensor scheduling , 2001, Syst. Control. Lett..

[8]  Michael Athans,et al.  1972 IFAC congress paper: On the determination of optimal costly measurement strategies for linear stochastic systems , 1972 .

[9]  Astrom.  Computer Controlled Systems , 1990 .

[10]  D. Kleinman,et al.  Optimal measurement scheduling for state estimation , 1992, [Proceedings] 1992 IEEE International Conference on Systems, Man, and Cybernetics.

[11]  J. Peschon,et al.  Optimal control of measurement subsystems , 1967, IEEE Transactions on Automatic Control.

[12]  A. Rantzer,et al.  Suboptimal dynamic programming with error bounds , 2002, Proceedings of the 41st IEEE Conference on Decision and Control, 2002..