This paper examines the problem of tracking control of networked multi-agent systems with multiple delays and impulsive effects, whose results are applied to mechanical robotic systems. Four kinds of impulsive effects are taken into account: 1) both the strengths of impulsive effects and the number of nodes injected with impulses are time dependent; 2) the strengths of impulsive effects occur according to certain probabilities and the number of nodes under impulsive control is time varying; 3) the strengths of impulses are time varying, whereas the number of nodes with impulses takes place according to certain probabilities; 4) both the strengths of impulses and the number of nodes with impulsive control occur according to certain probabilities. By utilizing the comparison principle, criteria are established for these different cases and a relationship between the frequencies (occurrence probabilities) of impulses and systems' parameters is unveiled. Finally, an example for tracking control of robotic systems is provided to show the effectiveness of the presented results.
Chengdu Section and York University, with media support of the Journal of Electronics & Information
This paper aims to investigate the problem of adaptive sliding mode control for Markov jump systems with deficient transition probability. Different from the existing literature, we propose a novel method to design a mode‐dependent sliding mode controller when the mode information is completely unknown, while the controller gain parameters can be designed by solving feasible conditions. Firstly, by designing an integral‐type sliding surface, on which an ideal sliding mode dynamics is obtained. Then, for different types of mode transition information, a set of feasible easy‐checking stochastic stability criteria are proposed for the sliding mode dynamics in terms of strict linear matrix inequalities. Further, relying on the parameters obtained from the stability criteria, an adaptive sliding mode controller is successfully designed with respect to different types of system mode transition information. Finally, a numerical example is provided to illustrate the advantage of the developed strategy.
No abstract is provided for this article.
No abstract is provided for this article.
In multi-autonomous underwater vehicle (multi-AUV) systems, the convergence rate is characterized by the pace of consistency of the key state information for each member. The topology with leader-follower architecture is designed as a combination of an undirected graph between followers and a digraph between leaders and followers. An overview of influences on convergence rate of the second-order consensus algorithm is elaborated in three aspects, along with the main contributions in this paper. Specifically, the explicit expression of the maximum convergence rate is established based on the root locus method, and then, the effects of control parameters on the convergence rate are analyzed. Moreover, the influences of network topologies on the convergence rate are investigated from the view of adjusting the existing connectivity, changing the weights on links, and utilizing hierarchical structure. The combination of consensus and filtering algorithm is also an approach to enhance the capacity of multi-AUV systems. In order to eliminate the accumulated errors in the process of dead reckoning, a collaborative navigation model is presented, and then, a localization approach based on consensus-unscented particle filter algorithm is proposed. Simulations results are provided to verify location performance under the assumption of Gaussian white noise in the systems. In addition, the influences of the topologies on positioning accuracy are explored.
A Mahalanobis hyperellipsoidal learning machine class incremental learning algorithm is proposed. To each class sample, the hyperellipsoidal that encloses as many as possible and pushes the outlier samples away is trained in the feature space. In the process of incremental learning, only one subclassifier is trained with the new class samples. The old models of the classifier are not influenced and can be reused. In the process of classification, considering the information of sample’s distribution in the feature space, the Mahalanobis distances from the sample mapping to the center of each hyperellipsoidal are used to decide the classified sample class. The experimental results show that the proposed method has higher classification precision and classification speed.
In this paper, an adaptive neural output-feedback tracking controller is designed for a class of multiple-input and multiple-output nonstrict-feedback nonlinear systems with time delay. The system coefficient and uncertain functions of our considered systems are both unknown. By employing neural networks to approximate the unknown function entries, and constructing a new input-driven filter, a backstepping design method of tracking controller is developed for the systems under consideration. The proposed controller can guarantee that all the signals in the closed-loop systems are ultimately bounded, and the time-varying target signal can be tracked within a small error as well. The main contributions of this paper lie in that the systems under consideration are more general, and an effective design procedure of output-feedback controller is developed for the considered systems, which is more applicable in practice. Simulation results demonstrate the efficiency of the proposed algorithm.
