Estimate accuracy versus measurement cost saving in continuous time linear filtering problems

We address the problem of defining a good trade off between reducing the cost of measurement and keeping the estimate accuracy within acceptable levels for continuous time linear system filtering problems. A measurement procedure which modifies the classical Kalman filter is proposed, allowing a sensible reduction of the measurement time intervals. This procedure is tested with satisfactory results against stable and unstable linear systems. © 2013 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.