Srdjan S. Stankovic

Integrated probabilistic data association

This paper presents an integrated probabilistic data association algorithm which provides recursive formulas for both data association and track quality (probability of track existence), allowing track initiation and track termination to be fully integrated into the association and smoothing algorithm. Integrated probabilistic data association is of similar computational complexity to probabilistic data association and as demonstrated by simulation, achieves comparable performance to the more computationally expensive interactive multiple model probabilistic data association algorithm which also integrates initiation and tracking. >

Consensus based overlapping decentralized estimation with missing observations and communication faults

In this paper a new algorithm for discrete-time overlapping decentralized state estimation of large scale systems is proposed in the form of a multi-agent network based on a combination of local estimators of Kalman filtering type and a dynamic consensus strategy, assuming intermittent observations and communication faults. Under general conditions concerning the agent resources and the network topology, conditions are derived for the convergence to zero of the estimation error mean and for the mean-square estimation error boundedness. A centralized strategy based on minimization of the steady-state mean-square estimation error is proposed for selection of the consensus gains; these gains can also be adjusted by local adaptation schemes. It is also demonstrated that there exists a connection between the network complexity and efficiency of denoising, i.e., of suppression of the measurement noise influence. Several numerical examples serve to illustrate characteristic properties of the proposed algorithm and to demonstrate its applicability to real problems.

Decentralized Parameter Estimation by Consensus Based Stochastic Approximation

In this paper, an algorithm for decentralized multi-agent estimation of parameters in linear discrete-time regression models is proposed in the form of a combination of local stochastic approximation algorithms and a global consensus strategy. An analysis of the asymptotic properties of the proposed algorithm is presented, taking into account both the multi-agent network structure and the probabilities of getting local measurements and implementing exchange of inter-agent messages. In the case of non-vanishing gains in the stochastic approximation algorithms, an asymptotic estimation error covariance matrix bound is defined as the solution of a Lyapunov-like matrix equation. In the case of asymptotically vanishing gains, the mean-square convergence is proved and the rate of convergence estimated. In the discussion, the problem of additive communication noise is treated in a methodologically consistent way. It is also demonstrated how the consensus scheme in the algorithm can contribute to the overall reduction of measurement noise influence. Some simulation results illustrate the obtained theoretical results.

Consensus Based Overlapping Decentralized Estimator

In this technical note a new algorithm for state estimation is proposed in the form of a multi-agent network based on a synergy between local Kalman filters and a dynamic consensus strategy between the agents. It is shown that it is possible, under general conditions concerning local resources and the network topology, to achieve asymptotic stability of the whole estimation algorithm by a proper choice of the consensus gains. It is demonstrated that the consensus gains can be obtained by minimizing the total mean-square estimation error. Capabilities of the network to achieve reduction of the measurement noise influence are also discussed.

Brief paper: Decentralized dynamic output feedback for robust stabilization of a class of nonlinear interconnected systems

The objective of this paper is to propose an approach to robust stabilization of systems which are composed of linear subsystems coupled by nonlinear time-varying interconnections satisfying quadratic constraints. The proposed algorithms, which are formulated within the convex optimization framework, employ linear dynamic feedback structure involving local Luenberger-type observers. It is also shown how the new methodology can produce improved results if interconnections have linear parts that are known a priori. Examples of output stabilization of inverted pendulums and decentralized control of a platoon of vehicles are used to illustrate the applicability of the design method.

An adaptive notch filter with improved tracking properties

An analysis of the properties of an adaptive notch filter (ANF) applied to time-varying frequency tracking is presented. Starting from the derivation of an expression for ANF output power, asymptotically optimal values for the pole contraction and forgetting factors are derived for recursive prediction error (RPE) type ANF algorithms. Based on the obtained results, a new ANF algorithm that includes adaptation of both pole contraction and forgetting factors is proposed. The given experimental results confirm the theoretical conclusions and show that the proposed algorithm is highly efficient in practice. >

Decentralized parameter estimation by consensus based stochastic approximation

In this paper, an algorithm for decentralized multi-agent estimation of parameters in linear discrete-time regression models is proposed in the form of a combination of local stochastic approximation algorithms and a global consensus strategy. An analysis of the asymptotic properties of the proposed algorithm is presented, taking into account both the multi-agent network structure and the probabilities of getting local measurements and implementing exchange of inter-agent messages. In the case of non-vanishing gains in the stochastic approximation algorithms, an asymptotic estimation error covariance matrix bound is defined as the solution of a Lyapunov-like matrix equation. In the case of asymptotically vanishing gains, the mean-square convergence is proved and the rate of convergence estimated. In the discussion, the problem of additive communication noise is treated in a methodologically consistent way. It is also demonstrated how the consensus scheme in the algorithm can contribute to the overall reduction of measurement noise influence. Some simulation results illustrate the obtained theoretical results.

Decomposition and decentralized control of systems with multi-overlapping structure

This paper considers decomposition and decentralized control of systems with multi-overlapping structure. It is demonstrated, using the inclusion principle, how the systems with longitudinal, loop and radial topologies can be expanded, and how the results can be used for designing controllers under information structure constraints. The proposed methodology is applied to automatic generation control (AGC) of an electric power system.

Stochastic inclusion principle applied to decentralized automatic generation control

An input-state-output inclusion principle for linear stochastic systems is proposed, with the emphasis on restriction and aggregation conditions for estimators and dynamic controllers. Inclusion of the LQG (linear quadratic Gaussian) optimal design is formulated and applied to the decentralized overlapping control of large-scale interconnected systems. Applications of the proposed methodology are illustrated using a stochastic model of automatic generation control (AGC) for interconnected power systems. Three types of overlapping decentralized and fully decentralized dynamic controllers, consisting of state estimators and feedback gains, are proposed for the cases of full and reduced measurement sets. An extensive analysis of both steady-state and transient regimes under a variety of operating conditions shows the superiority of the proposed AGC scheme with respect to the standard AGC designs.

Robust stabilization of nonlinear interconnected systems by decentralized dynamic output feedback

The objective of this note is to propose a dynamic output control scheme within the LMI framework for robust decentralized stabilization of systems composed of linear dynamic subsystems coupled by static nonlinear interconnections satisfying quadratic constraints. The procedure utilizes the general linear dynamic feedback structure, and consists of two steps, the first giving a block-diagonal Lyapunov matrix together with the robustness degree, and the second the controller parameters. A numerical example illustrates the applicability of the method.