Distributed Fusion With Unknown Inputs Under Bandwidth-Aware Event-Triggered Mechanisms: Monotonicity and Boundedness

This article addresses the distributed fusion estimation problem for a class of nonlinear systems subject to both unknown inputs and sensor failures, where a bandwidth-aware dynamic event-triggered strategy is proposed to reduce communication burden and energy consumption. First, a novel distributed filter is constructed with the aid of an intermediate variable resolving unknown inputs. In light of such a structure combined with Kalman filtering theory, an upper bound of the filtering error covariance is derived and subsequently minimized by designing appropriate gains. Furthermore, the distributed fusion depends on an optimization obtained under the covariance intersection fusion strategy. The ultimate boundedness and the monotonicity with respect to triggered thresholds are discussed for the minimized upper bound of the fusing error covariance. Finally, the effectiveness of the proposed method is vivid through illustrative simulation examples.