Fusion estimation for multi-rate linear repetitive processes under weighted try-once-discard protocol

In this paper, the fusion estimation problem is studied for a class of discrete time-varying multi-rate linear repetitive processes (LRPs) under weighted try-once-discard protocol. The LRPs are measured by multiple sensors that are allowed to have different sampling periods, and the state updating period of the LRPs is also allowed to be different from the sampling periods of the asynchronous sensors. To facilitate the estimator design, the lifting technique is applied to transform the multi-rate LRPs to single-rate ones. Moreover, due to limited communication capability, the weighted try-once-discard protocol is adopted to schedule the asynchronous sensors. A set of local estimators is designed such that the upper bounds on the local estimation error covariances are guaranteed, and such upper bounds are then minimized by appropriately designing the estimator gains. Furthermore, the estimates from the local estimators are fused by recurring to the sequential covariance intersection fusion method. Finally, a simulation example is given to demonstrate the effectiveness of the proposed fusion estimation scheme.