Maximum-correntropy-based Kalman filtering for time-varying systems with randomly occurring uncertainties: An event-triggered approach

In this article, the maximum-correntropy-based Kalman filtering problem is investigated for a class of linear time-varying systems in the presence of non-Gaussian noises and randomly occurring uncertainties (ROUs). The random nature of the parameter uncertainties is characterized by a stochastic variable conforming to the Bernoulli distribution. In order to avoid unnecessary data transmission and reduce consumption of limited communication resource, the event-triggered mechanism (ETM) is introduced in the sensor-to-filter channel to decide whether the data should be transmitted or not. A novel performance index is first proposed to reflect the joint effects from the non-Gaussian noises, the ETM as well as the ROUs. Under the proposed performance index, an event-based Kalman filter is then constructed whose gain is calculated based on the maximum correntropy criterion. Finally, the effectiveness of the proposed filtering scheme is verified via a practical target tracking example.

Automated summary

This article investigates a Kalman filtering algorithm based on maximum correntropy for linear time-varying systems with non-Gaussian noises and randomly occurring uncertainties. The event-triggered mechanism is introduced to reduce unnecessary data transmission and communication resource consumption, and a novel performance index is proposed to reflect the joint effects from various factors.