Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

This paper provides a theoretical framework for analysis of consensus algorithms for multi-agent networked systems with an emphasis on the role of directed information flow, robustness to changes in network topology due to link/node failures, time-delays, and performance guarantees. An overview of basic concepts of information consensus in networks and methods of convergence and performance analysis for the algorithms are provided. Our analysis framework is based on tools from matrix theory, algebraic graph theory, and control theory. We discuss the connections between consensus problems in networked dynamic systems and diverse applications including synchronization of coupled oscillators, flocking, formation control, fast consensus in small-world networks, Markov processes and gossip-based algorithms, load balancing in networks, rendezvous in space, distributed sensor fusion in sensor networks, and belief propagation. We establish direct connections between spectral and structural properties of complex networks and the speed of information diffusion of consensus algorithms. A brief introduction is provided on networked systems with nonlocal information flow that are considerably faster than distributed systems with lattice-type nearest neighbor interactions. Simulation results are presented that demonstrate the role of small-world effects on the speed of consensus algorithms and cooperative control of multivehicle formations

/pdf/consensus-and-cooperation-in-networked-multi-agent-systems-5ame7z72ny.pdf

Consensus and Cooperation in Networked Multi-Agent Systems

Data Mining Practical Machine Learning Tools and Techniques

This paper reviews some main results and progress in distributed multi-agent coordination, focusing on papers published in major control systems and robotics journals since 2006. Distributed coordination of multiple vehicles, including unmanned aerial vehicles, unmanned ground vehicles, and unmanned underwater vehicles, has been a very active research subject studied extensively by the systems and control community. The recent results in this area are categorized into several directions, such as consensus, formation control, optimization, and estimation. After the review, a short discussion section is included to summarize the existing research and to propose several promising research directions along with some open problems that are deemed important for further investigations.

/pdf/an-overview-of-recent-progress-in-the-study-of-distributed-1xnb6e12se.pdf

An Overview of Recent Progress in the Study of Distributed Multi-Agent Coordination

System Identification I

We study feedback control for a discrete-time integrator with unitary delay in the presence of quantization both in the control action and in the measurement of the controlled variable. In some applications the quantization effects can be neglected, but when high precision is needed, they have to be explicitly accounted for in control design. In this paper, we propose a switched control solution for minimizing the effect of quantization of both the control and controlled variables for the considered system, that is quite common in the computing systems domain, for example, in thread scheduling, clock synchronization, and resource allocation. We show that the switched solution outperforms the one without switching, designed by neglecting quantization, and analyze necessary and sufficient conditions for the controlled system to exhibit periodic solutions in the presence of an additive constant disturbance affecting the control input. Simulation results provide evidence of the effectiveness of the approach.

Switched Control for Quantized Feedback Systems: Invariance and Limit Cycle Analysis

In this paper, we propose a semantic indexing algorithm based on the controlled Markov chain modeling framework. Controlled Markov chain models are used to describe the temporal evolution of low-level visual descriptors extracted from the MPEG compressed bit-stream. To reduce the number of false detections given by the proposed video-processing algorithm, we have considered also the audio signal. In particular we have evaluated the "loudeness" associated to each video segments identified by the analysis carried out on the video signal. The intensity of the "loudness" has then been used to order the selected video segments. In this way, the segments associated to interesting events appear in the very first positions of the ordered list, and the number of false detections can be greatly reduced. The proposed algorithm has been conceived for soccer game video sequences, and the simulation results have shown the effectiveness of the proposed algorithm.

/pdf/a-markov-chain-model-for-semantic-indexing-of-sport-program-125fgu1k8d.pdf

A Markov chain model for semantic indexing of sport program sequences.

In this paper, we present a new approach to control design in presence of constraints. This approach relies on the reformulation of the controller design problem as a semi-infinite convex optimization program, and on the solution of this program by the scenario optimization technology. The approach is illustrated through a simple example of disturbance rejection subject to input saturation constraints.

/pdf/a-new-approach-to-controller-design-in-presence-of-44tdmhegy8.pdf

A new approach to controller design in presence of constraints

This paper studies the problem of characterizing the region of the airspace that will be occupied by a space debris during an uncontrolled reentry ( footprint ), with the final goal of supporting the air traffic controllers in their task of guiding aircraft safely from their origin to their destination. Given the various sources of uncertainty affecting the debris dynamics, the reentry process is characterized probabilistically and the problem of determining the footprint is formulated in terms of a chance-constrained optimization program, which is solved via a simulation-based method. When observations of the debris initial position and radar measurements of the aircraft prior to the reentry event are available, nonlinear filtering techniques can be adopted and the posterior probability distribution of the debris position as well as of the wind field affecting the reentry process can be integrated in the chance-constraint formulation so as to obtain an enhanced estimate of the footprint. Simulation results show the efficacy of the approach.

/pdf/a-randomized-approach-to-probabilistic-footprint-estimation-598gvk4104.pdf

A Randomized Approach to Probabilistic Footprint Estimation of a Space Debris Uncontrolled Reentry

We consider the problem of shaping the distribution of the state of a discrete time linear system subject to a stationary disturbance so as to optimize a performance index while satisfying probabilistic constraints involving control input and state. The state is not available and a disturbance compensator is designed via a one-shot off-line computational approach to optimally operate the system in stationary conditions. The resulting control strategy is easy to implement, able to handle probabilistic constraints and guarantees optimality in the long run. Moreover, it does not require the on-line (re)computation of the control parameters. Interestingly, problems where the disturbance is cyclostationary and/or the control is periodic can be embedded in our formulation. This is the case of the numerical example presented in the paper, where the proposed methodology is applied to stochastic periodic control of a battery for peak shaving of the solar energy produced by a photovoltaic panel installation.

/pdf/optimally-shaping-the-stationary-distribution-of-a-38n170rhpl.pdf

Maria Prandini

Papers

Switched Control for Quantized Feedback Systems: Invariance and Limit Cycle Analysis

A Markov chain model for semantic indexing of sport program sequences.

A new approach to controller design in presence of constraints

A Randomized Approach to Probabilistic Footprint Estimation of a Space Debris Uncontrolled Reentry

Optimally shaping the stationary distribution of a constrained discrete time stochastic linear system via disturbance compensation