This article is concerned with a distributed filtering problem for Markov jump systems subject to the measurement loss with unknown probabilities. A centralized robust Kalman filter is designed by using variational Bayesian methods and a modified interacting multiple model method based on information theory (IT-IMM). Then, a distributed robust Kalman filter based on the centralized filter and a hybrid consensus method called hybrid consensus on measurement and information (HCMCI) is designed. Moreover, boundedness of the estimation errors and the estimation error covariances are studied for the distributed robust Kalman filter.
This paper aims to solve the event-based output consensus problem for multiagent systems on directed graphs in the presence of denial-of-service (DoS) attacks. The proposed protocol relies on a node-based self-triggered mechanism that can avoid continuous communication and the use of a strongly connected graph. This means that neither the controller nor the triggering function relies on continuous communication. By exploiting the output regulation strategy, the controller can realize output resilient consensus for linear heterogeneous multiagent systems against DoS attacks using relative observed information. This protocol offers a significant benefit in that it can be applied to directed networks without requiring the Laplacian matrix information to be taken into account. Moreover, another superiority lies in that it can handle with the situation that the communication graph contains link faults. Finally, through numerical simulations, the effectiveness of the designed protocol is demonstrated.
7 July 2020 in IEEE Transactions on Systems, Man, and Cybernetics: Systems
The sequential fusion estimation for multirate multisensor dynamic systems with heavy-tailed noises and unreliable measurements is an important problem in dynamic system control. This work proposes a sequential fusion algorithm and a detection technique based on Student's t-distribution and the approximate t-filter. The performance of the proposed algorithm is analyzed and compared with the Gaussian Kalman filter-based sequential fusion and the t-filter-based sequential fusion without detection technique. Theoretical analysis and exhaustive experimental analysis show that the proposed algorithm is effective and robust to unreliable measurements. The t-filter-based sequential fusion algorithm is shown to be the generalization of the classical Gaussian Kalman filter-based optimal sequential fusion algorithm.
8 April 2024 in IEEE Transactions on Circuits and Systems I: Regular Papers
As a critical security threat, Byzantine attacks in nodes sending tampered data to their neighbors in networks, significantly undermining the robustness of networked systems. This paper focuses on the leader-follower resilient consensus of continuous-time heterogeneous multiagent systems against Byzantine attacks on nodes. Considering the excessive communication burden in practice, a distributed dynamic event-triggered mechanism is devised to enable intermittent communication among agents. Building upon this mechanism, two protocols are applied to achieve the output consensus when the leader converges to dynamic and static states, respectively. The latter protocol can decrease both communication data and the frequency of event triggers when the leader converges to a static state. In addition, the output regulation control for uncertain system dynamics is achieved by utilizing the technique of internal model principle. Moreover, the designed event-triggered mechanism can be employed to the networks with bounded communication delays. Finally, the effectiveness of the designed protocols is demonstrated by some numerical simulations.
7 November 2022 in IEEE Transactions on Aerospace and Electronic Systems
In this paper, a distributed filtering problem is studied for a Markov jump system over sensor networks, where measurements are partially disturbed by outliers. A local multiple model filter is designed based on variational Bayesian approaches and interacting multiple model methods (IMM), the designed filter is able to identify and exclude outliers automatically, so as to mitigate the impact of outliers. A distributed filter is proposed by combining the designed local filter with consensus on information methods. Furthermore, a sufficient condition is given to guarantee the stability of the designed distributed filter, in which the estimation errors of each sensor are bounded in the mean square sense. Finally, both simulations and experiments of target tracking systems are done to show the effectiveness of the designed distributed filter.