Showing papers in "arXiv: Optimization and Control in 2003"

Coordination and geometric optimization via distributed dynamical systems

Abstract: This paper discusses dynamical systems for disk-covering and sphere-packing problems. We present facility location functions from geometric optimization and characterize their differentiable properties. We design and analyze a collection of distributed control laws that are related to nonsmooth gradient systems. The resulting dynamical systems promise to be of use in coordination problems for networked robots; in this setting the distributed control laws correspond to local interactions between the robots. The technical approach relies on concepts from computational geometry, nonsmooth analysis, and the dynamical system approach to algorithms.

Linear Matrix Inequality Representation of Sets

Abstract: This article concerns the question: which subsets of ${\mathbb R}^m$ can be represented with Linear Matrix Inequalities, LMIs? This gives some perspective on the scope and limitations of one of the most powerful techniques commonly used in control theory. Also before having much hope of representing engineering problems as LMIs by automatic methods one needs a good idea of which problems can and cannot be represented by LMIs. Little is currently known about such problems. In this article we give a necessary condition, we call "rigid convexity", which must hold for a set ${\cC} \in {\mathbb R}^m$ in order for ${\cC}$ to have an LMI representation. Rigid convexity is proved to be necessary and sufficient when $m=2$. This settles a question formally stated by Pablo Parrilo and Berndt Sturmfels in [PSprep].

The Convergence of the Extended Kalman Filter

Abstract: We demonstrate that the extended Kalman filter converges locally for a broad class of nonlinear systems If the initial estimation error of the filter is not too large then the error goes to zero exponentially as time goes to infinity To demonstrate this, we require that the system be C 2 and uniformly observable with bounded second partial derivatives

A theory of bond portfolios

Abstract: We introduce a bond portfolio management theory based on foundations similar to those of stock portfolio management. A general continuous-time zero-coupon market is considered. The problem of optimal portfolios of zero-coupon bonds is solved for general utility functions, under a condition of no-arbitrage in the zero-coupon market. A mutual fund theorem is proved, in the case of deterministic volatilities. Explicit expressions are given for the optimal solutions for several utility functions.

The control transmutation method and the cost of fast controls

Abstract: In this paper, the null controllability in any positive time T of the first-order equation (1) x'(t)=e^{i\theta}Ax(t)+Bu(t) (|\theta|<\pi/2 fixed) is deduced from the null controllability in some positive time L of the second-order equation (2) z''(t)=Az(t)+Bv(t). The differential equations (1) and (2) are set in a Banach space, B is an admissible unbounded control operator, and A is a generator of cosine operator function. The control transmutation method explicits the input function u of (1) in terms of the input function v of (2): u(t,x)=\int k(t,s)v(s)ds, where the compactly supported kernel k depends on T and L only. It proves that the norm of a u steering the system (1) from an initial state x_{0} to zero grows at most like ||x_{0}||\exp(\alpha_{*}L^{2}/T) as the control time T tends to zero. (The rate \alpha_{*} is characterized independently by a one-dimensional controllability problem.) In the applications to the cost of fast controls for the heat equation, L is the length of the longest ray of geometric optics which does not intersect the control region.

Convolutional Codes with Maximum Distance Profile

Abstract: Maximum distance profile codes are characterized by the property that two trajectories which start at the same state and proceed to a different state will have the maximum possible distance from each other relative to any other convolutional code of the same rate and degree. In this paper we use methods from systems theory to characterize maximum distance profile codes algebraically. Tha main result shows that maximum distance profile codes form a generic set inside the variety which parametrizes the set of convolutional codes of a fixed rate and a fixed degree.

Static Arbitrage Bounds on Basket Option Prices

Abstract: We consider the problem of computing upper and lower bounds on the price of a European basket call option, given prices on other similar baskets. Although this problem is very hard to solve exactly in the general case, we show that in some instances the upper and lower bounds can be computed via simple closed-form expressions, or linear programs. We also introduce an efficient linear programming relaxation of the general problem based on an integral transform interpretation of the call price function. We show that this relaxation is tight in some of the special cases examined before.

Flexible Complementarity Solvers for Large-Scale Applications

Abstract: Discretizations of infinite-dimensional variational inequalities lead to linear and nonlinear complementarity problems with many degrees of freedom. To solve these problems in a parallel computing environment, we propose two active-set methods that solve only one linear system of equations per iteration. The linear solver, preconditioner, and matrix structures can be chosen by the user for a particular application to achieve high parallel performance. The parallel scalability of these methods is demonstrated for some discretizations of infinite-dimensional variational inequalities.

Quasi-Invariant Optimal Control Problems

Abstract: We study in optimal control the important relation between invariance of the problem under a family of transformations, and the existence of preserved quantities along the Pontryagin extremals. Several extensions of Noether theorem are provided, in the direction which enlarges the scope of its application. We formulate a more general version of Noether's theorem for optimal control problems, which incorporates the possibility to consider a family of transformations depending on several parameters and, what is more important, to deal with quasi-invariant and not necessarily invariant optimal control problems. We trust that this latter extension provides new possibilities and we illustrate it with several examples, not covered by the previous known optimal control versions of Noether's theorem.

Comparison of the methods for discrete approximation of the fractional-order operator

Abstract: Porovnanie metod pre diskretnu aproximaciu operatorov necelociselneho radu In this paper we will present some alternative types of discretization methods (discrete approximation) for the fractional-order (FO) differentiator and their application to the FO dynamical system described by the FO differential equation (FDE). With analytical solution and numerical solution by power series expansion (PSE) method are compared two effective methods - the Muir expansion of the Tustin operator and continued fraction expansion method (CFE) with the Tustin operator and the Al-Alaoui operator. Except detailed mathematical description presented are also simulation results. From the Bode plots of the FO differentiator and FDE and from the solution in the time domain we can see, that the CFE is a more effective method according to the PSE method, but there are some restrictions for the choice of the time step. The Muir expansion is almost unusable.

Digital Fractional Order Controllers Realized by PIC Microprocessor: Experimental Results

Abstract: This paper deals with the fractional-order controllers and their possible hardware realization based on PIC microprocessor and numerical algorithm coded in PIC Basic. The mathematical description of the digital fractional -order controllers and approximation in the discrete domain are presented. An example of realization of the particular case of digital fractional-order PID controller is shown and described.

Gauge Symmetries And Noether Currents In Optimal Control

Abstract: We extend the second Noether theorem to optimal control problems which are invariant under symmetries depending upon k arbitrary functions of the indepen- dent variable and their derivatives up to some order m. As far as we consider a semi-invariance notion, and the trans formation group may also depend on the control variables, the result is new even in the classical context of the calculus of variations.

Efficient Determination of Gibbs Estimators with Submodular Energy Functions

Abstract: Now the Gibbs estimators are used to solve many problems in computer vision, optimal control, Bayesian statistics etc. Their determination is equivalent to minimization of corresponding energy functions. Therefore, the real use of the estimators is restricted by opportunity to compute or evaluate satisfactory values of the energy functions. New combinatorial techniques that have been developed for the last years allowed efficient employment of high dimensional Ising models with Boolean and rational valued variables. In addition, graph cut methods were proposed to minimize energy functions U = P gµ(xµ)+ P gµ,�(xµ, x�)+ P gµ,�,�(xµ, x�, x�) of Boolean variables xµ and U = P µ h(iµ)+ P µ,� g(|iµ −j�|) of integer variables iµ with submodular functions g. In the paper energy functions that often occur in applications (for instance, in estimating of images) are considered. They are suppose to be depended on variables taking values in totally ordered finite sets (finite sets of arbitrary numbers are among them). The representation of functions by Boolean polynomials are described. Fast methods of minimization of submodular Boolean polynomials that correspond to the energy functions are developed. For that the Modified algorithm of minimization of submodular functions (MSFM) is proposed. Also the minimum graph cut technique is presented for submodular Boolean polynomials that satisfy some additional condition. Among such polynomials are, for instance, those with all coefficients before nonlinear monomials negative as well as submodular Boolean polynomials, which have all coefficients before monomials of order more than two positive. Mathematics Subject Classification 2000: 90C27, 62N, 62H, 68W

Convolutional Codes of Goppa Type

Abstract: A new kind of Convolutional Codes generalizing Goppa Codes is proposed. This provides a systematic method for constructing convolutional codes with prefixed properties. In particular, examples of Maximum-Distance Separable (MDS) convolutional codes are obtained.

The stochastic goodwill problem

Abstract: Stochastic control problems related to optimal advertising under uncertainty are considered. In particular, we determine the optimal strategies for the problem of maximizing the utility of goodwill at launch time and minimizing the disutility of a stream of advertising costs that extends until the launch time for some classes of stochastic perturbations of the classical Nerlove-Arrow dynamics. We also consider some generalizations such as problems with constrained budget and with discretionary launching.

Time-dependent unidirectional communication in multi-agent systems

Abstract: We study a simple but compelling model of $n$ interacting agents via time-dependent, unidirectional communication. The model finds wide application in a variety of fields including synchronization, swarming and distributed decision making. In the model, each agent updates his current state based upon the current information received from other agents. Necessary and/or sufficient conditions for the convergence of the individual agents' states to a common value are presented, extending recent results reported in the literature. Unlike previous, related studies, the approach of the present paper does not rely on algebraic graph theory and is of a completely nonlinear nature. It is rather surprising that with these nonlinear tools, extensions may be obtained even for the linear cases discussed in the literature. The proof technique consists of a blend of graph-theoretic and system-theoretic tools integrated within a formal framework of set-valued Lyapunov theory and may be of independent interest. Further, it is also observed that more communication does not necessarily lead to better convergence and may eventually even lead to a loss of convergence, even for the simple models discussed in the present paper.

A Semidefinite Representation for some Minimum Cardinality Problems

Abstract: Using techniques developed in [Lasserre02], we show that some minimum cardinality problems subject to linear inequalities can be represented as finite sequences of semidefinite programs. In particular, we provide a semidefinite representation of the minimum rank problem on positive semidefinite matrices. We also use this technique to cast the problem of finding convex lower bounds on the objective as a semidefinite program.

Algebraic Recipes for Integer Programming

Abstract: Integer programming is concerned with solving linear systems of equations over the non-negative integers. The basic question is to find a solution which minimizes a given linear objective function for a fixed right hand side. Here we also consider parametric versions where the objective function and the right hand side are allowed to vary. The main emphasis is on Gr\"obner bases, rational generating functions, and how to use existing software packages. Concrete applications to problems in statistical modeling will be presented. These are the notes for a lecture to be given during the AMS Short Course, ``Trends in Optimization'', prior to the Joint Mathematics Meetings in Phoenix on January 5-6, 2004.

Optimization of light structures: the vanishing mass conjecture

Abstract: We consider the shape optimization problem which consists in placing a given mass $m$ of elastic material in a design region so that the compliance is minimal. Having in mind optimal light structures, our purpose is to show that the problem of finding the stiffest shape configuration simplifies as the total mass $m$ tends to zero: we propose an explicit relaxed formulation where the compliance appears after rescaling as a convex functional of the relative density of mass. This allows us to write necessary and sufficient optimality conditions for light structures following the Monge-Kantorovich approach developed recently in [5].

Positive forward rates in the maximum smoothness framework

Abstract: In this article we present a non-linear dynamic programming algorithm for the computation of forward rates within the maximum smoothness framework. The algorithm implements the forward rate positivity constraint for a one-parametric family of smoothness measures and it handles price spreads in the constraining dataset. We investigate the outcome of the algorithm using the Swedish Bond market showing examples where the absence of the positive constraint leads to negative interest rates. Furthermore we investigate the predictive accuracy of the algorithm as we move along the family of smoothness measures. Among other things we observe that the inclusion of spreads not only improves the smoothness of forward curves but also significantly reduces the predictive error.

The Lax conjecture is true

Abstract: In 1958 Lax conjectured that hyperbolic polynomials in three variables are determinants of linear combinations of three symmetric matrices. This conjecture is equivalent to a recent observation of Helton and Vinnikov.

Further Results on Lyapunov Functions and Domains of Attraction for Perturbed Asymptotically Stable Systems

Abstract: We present new theorems characterizing robust Lyapunov functions and infinite horizon value functions in optimal control as unique viscosity solutions of partial differential equations. We use these results to further extend Zubov's method for representing domains of attraction in terms of partial differential equation solutions.

Optimal Convergence Trading

Abstract: This article examines arbitrage investment in a mispriced asset when the mispricing follows the Ornstein-Uhlenbeck process and a credit-constrained investor maximizes a generalization of the Kelly criterion. The optimal differentiable and threshold policies are derived. The optimal differentiable policy is linear with respect to mispricing and risk-free in the long run. The optimal threshold policy calls for investing immediately when the mispricing is greater than zero with the investment amount inversely proportional to the risk aversion parameter. The investment is risky even in the long run. The results are consistent with the belief that credit-constrained arbitrageurs should be risk-neutral if they are to engage in convergence trading.

Convolutional Goppa Codes

Abstract: We define Convolutional Goppa Codes over algebraic curves and construct their corresponding dual codes. Examples over the projective line and over elliptic curves are described, obtaining in particular some Maximum-Distance Separable (MDS) convolutional codes.

Uniform stability of damped nonlinear vibrations of an elastic string

Abstract: Here we are concerned about uniform stability of damped nonlinear transverse vibrations of an elastic string fixed at its two ends. The vibrations governed by nonlinear integro-differential equation of Kirchoff type, is shown to possess energy uniformly bounded by exponentially decaying function of time. The result is achieved by considering an energy-like Lyapunov functional for the system.

Matrosov's theorem using a family of auxiliary functions: an analysis tool to aid time-varying nonlinear control design

Abstract: We present a new result on uniform attractivity of the origin for nonlinear time-varying systems. Our theorem generalizes Matrosov's theorem which extends, in a certain manner, Krasovskii-LaSalle invariance principle to the case of general nonlinear time-varying systems. We show the utility of our theorem by addressing a control problem of port interconnected driftless systems. The latter includes as special case, the control of chained-form nonholonomic systems which has been extensively studied in the literature.

The structured distance to ill-posedness for conic systems

Abstract: An important measure of conditioning of a conic linear system is the size of the smallest structured perturbation making the system ill-posed. We show that this measure is unchanged if we restrict to perturbations of low rank. We thereby derive a broad generalization of the classical Eckart-Young result characterizing the distance to ill-posedness for a linear map.

Integrals of Motion for Discrete-Time Optimal Control Problems

Abstract: We obtain a discrete time analog of E. Noether's theorem in Optimal Control, asserting that integrals of motion associated to the discrete time Pontryagin Maximum Principle can be computed from the quasi-invariance properties of the discrete time Lagrangian and discrete time control system. As corollaries, results for first-order and higher-order discrete problems of the calculus of variations are obtained.

Nonlinear internal models for output regulation

Abstract: In this paper we show how nonlinear internal models can be effectively used in the design of output regulators for nonlinear systems. This result provides a significant enhancement of the non-equilibrium theory for output regulation, which we have presented in the recent paper entitled "Limit Sets, Zero Dynamics, and Internal Models in the Problem of Nonlinear Output Regulation".

Computing the Integer Programming Gap

Abstract: We determine the maximal gap between the optimal values of an integer program and its linear programming relaxation, where the matrix and cost function are fixed but the right hand side is unspecified. Our formula involves irreducible decomposition of monomial ideals. The gap can be computed in polynomial time when the dimension is fixed.