Remarks on lower semicontinuity and lower closure

In the context of a recent geometric condition of Cesari, used in the reduction of seminormality requirements in lower closure theorems, this paper shows that the existence of a strongly convergent selection from the sequence of orientor fields, under Kuratowski property (K), is adequate to guarantee lower closure theorems. This generalization is justified through examples. Several related remarks are made.This work was done in the framework of Research Project AFOSR-71-2122 at the University of Michigan, Ann Arbor, Michigan. The author wishes to thank Professor L. Cesari for helpful discussions during the writing of this paper.     

Pontryagin functions for multiple integral control problems

An optimal control problem is considered whose performance index is represented by anm-fold multiple integral, the state equations being given by a system of first-order partial differential equations. The concept of a field of optimal control variables with respect to an independent integral is introduced, the significance of these fields being due to the fact that a control pair is optimal whenever it is imbedded in such a field. Since there arem distinctm-fold independent integrals, it is possible to constructm distinct fields of this kind. For each of these, a Pontryagin function is defined, and it is shown that, if an optimal pair is embedded in one of these fields, it satisfies a corresponding Pontryagin maximum principle.This research was supported in part by NSF Grant No. GP-32830     

Lower closure and existence theorems for optimal control problems with infinite horizon

Lower closure theorems are proved for optimal control problems governed by ordinary differential equations for which the interval of definition may be unbounded. One theorem assumes that Cesari's property (Q) holds. Two theorems are proved which do not require property (Q), but assume either a generalized Lipschitz condition or a bound on the controls in an appropriateL _p-space. An example shows that these hypotheses can hold without property (Q) holding.     

Sufficiency conditions for multiple integral control problems

The construction of multiple integrals which are independent, in the sense that they depend solely on the values of their integrands on the boundary of the domain of integration is described. These integrals are applied to the derivation of a sufficiency condition for multiple integral optimal control problems.This research was supported in part by NSF Grant No. GP-32830.     

Nemitsky's operators and lower closure theorems

This paper focuses on certain analytic criteria given by the authors in earlier works, for the geometric property of upper semicontinuity of set-valued functions, used in the proofs of lower closure theorems, and hence in existence theory. In particular, it is observed that, under Filippov-type condition (namely, when the set of controls is bounded in measure or in norm), mere Carathéodory-type continuity of the relevant functionsf is sufficient to guarantee a weak form of property (Q), and in turn the lower closure theorems.This work has been partially supported by Research Project AFOSR-71-2122 at the University of Michigan, Ann Arbor, Michigan.     

Geometric methods for nonlinear optimal control problems

It is the purpose of this paper to develop and present new approaches to optimal control problems for which the state evolution equation is nonlinear. For bilinear systems in which the evolution equation is right invariant, it is possible to use ideas from differential geometry and Lie theory to obtain explicit closed-form solutions.The author wishes to thank Professor A. Krener for many stimulating discussions and in particular for suggesting Theorem 3.3. Also, special thanks are due to the author's thesis advisor Professor R. W. Brockett under whose direction most of the research was done. Finally, the author thanks two anonymous referees for suggestions which have improved the exposition.     

An existence theorem in the calculus of variations

In this paper, a nonparametric variational problem is considered in the setting of the theory of generalized curves. It is assumed that the integrand of the problem does not grow at infinity faster than the norm of the variable , for all values of the other variablest andx (which take their values in a compact product set). It is shown that a generalized curve exists such that the minimum of the functional over an appropriate set is achieved. This generalized curve does not in general have compact support.     

A note on the use of direct sufficient conditions in optimal control problems

It is always possible to transform a nonautonomous optimal control problem into an autonomous one. However, the direct sufficient conditions may yield no information when applied to this autonomous problem, even though they do allow one to conclude sufficiency when applied to the original nonautonomous problem.This research was supported by the Air Force Office of Scientific Research, under Grant No. AFOSR-76-2923.     

A useful approximation scheme for Lagrangians

We present a new, useful approximation scheme for the integrand of an integral functional, revolving around a generalized bipolarity result. This scheme leads immediately to lower semicontinuity and lower closure results for the integral functional, as well as to other, more general seminormality properties.     

Hamilton-Jacobi theory for periodic control problems

The general problem of periodic optimization is considered in this paper. The contribution consists in stating sufficient conditions for the optimality of a control satisfying the periodicity constraint. The result has been achieved by means of the classical Hamilton-Jacobi equation, suitably modified in order to consider the peculiar constraint of the problem. Finally, an interesting application of the theory to the case of linear, time-invariant systems is given.This work was supported by CNR (Consiglio Nazionale delle Ricerche), Rome, Italy.     

A lagrange multiplier theorem for control problems with state constraints

We prove an existence theorem of Lagrange multipliers for an abstract control problem in Banach spaces. This theorem may be applied to obtain optimality conditions for control problems governed by partial differential equations in the presence of pointwise state constraints.     

A theorem of Briançon-Skoda type for regular local rings containing a field

Let be a regular local ring containing a field. We give a refinement of the Briançon-Skoda theorem showing that if is a minimal reduction of where is -primary, then where and is the largest ideal such that . The proof uses tight closure in characteristic and reduction to characteristic for rings containing the rationals.

     

Interior variations in dynamic optimization problems

In this article, the author examines the properties of interior variations and indicates how to use them in order to formulate the necessary condition of optimality for problems of dynamic optimization, in particular, problems of variational calculus and of optimal control. For optimal control problems, an optimization technique based on interior variations and polynomial approximations is suggested and then illustrated by an explanatory example.     

A saddle-point criterion for convex problems with infinite-dimensional equality constraints

In this paper, we develop a saddle-point criterion for convex optimization problems with infinite-dimensional equality constraints. The method used in the derivation of this criterion is based on the property of openness of the equality operator. As an application, we develop necessary and sufficient conditions for an optimal control problem under a suitable controllability assumption. Constructing an optimal solution for a production planning problem is used as an illustrative example.The authors would like to thank Professor R. T. Rockafellar for his suggestions which have resulted in an improved version of the paper.     

Sufficient conditions for a minimum of the free-final-time optimal control problem

The sufficient conditions for a minimum of the free-final-time optimal control problem are the strengthened Legendre-Clebsch condition and the conjugate point condition. In this paper, a new approach for determining the location of the conjugate point is presented. The sweep method is used to solve the linear two-point boundary-value problem for the neighboring extremal path from a perturbed initial point to the final constraint manifold. The new approach is to solve for the final condition Lagrange multiplier perturbation and the final time perturbation simultaneously. Then, the resulting neighboring extremal control is used to write the second variation as a perfect square and obtain the conjugate point condition. Finally, two example problems are solved to illustrate the application of the sufficient conditions.     

Formulation of Euler-Lagrange equations for fractional variational problems

This paper presents extensions to traditional calculus of variations for systems containing fractional derivatives. The fractional derivative is described in the Riemann-Liouville sense. Specifically, we consider two problems, the simplest fractional variational problem and the fractional variational problem of Lagrange. Results of the first problem are extended to problems containing multiple fractional derivatives and unknown functions. For the second problem, we also present a Lagrange type multiplier rule. For both problems, we develop the Euler-Lagrange type necessary conditions which must be satisfied for the given functional to be extremum. Two problems are considered to demonstrate the application of the formulation. The formulation presented and the resulting equations are very similar to those that appear in the field of classical calculus of variations.     

General necessary conditions for infinite horizon fractional variational problems

We consider infinite horizon fractional variational problems, where the fractional derivative is defined in the sense of Caputo. Necessary optimality conditions for higher-order variational problems and optimal control problems are obtained. Transversality conditions are obtained in the case state functions are free at the initial time.     

First-order necessary conditions for infinite-horizon variational problems

We establish rigorously several pointwise or asymptotic firstorder necessary conditions for infinite-horizon variational problems in general form, in the framework of continuous time. We obtain several new results, and we extend to general differentiable Lagrangians some results known only in special cases. To realize this aim, we justify two different ways to associate a family of finite-horizon problems to an infinite-horizon problem.The authors thank an anonymous referee for providing important historical references     

Necessary conditions for problems with higher derivative bounded state variables

This paper combines the separate works of two authors. Tan proves a set of necessary conditions for a control problem with second-order state inequality constraints (see Ref. 1). Russak proves necessary conditions for an extended version of that problem. Specifically, the extended version augments the original problem by including state equality constraints, differential and isopermetric equality and inequality constraints, and endpoint constraints. In addition, Russak (i) relaxes the solvability assumption on the state constraints, (ii) extends the maximum principle to a larger set, (iii) obtains modified forms of the relationH =H _t and of the transversality relation usually obtained in problems of this type, and (iv) proves a condition concerning (t ¹), the derivative of the multiplier functions at the final time.Russak's work was supported by a NPS Foundation Grant.Tan is indebted to his thesis advisor, Professor M. R. Hestenes, for suggesting the topic and for his help and guidance in the development of his work. Tan's work was supported by the Army Research Office, Contract No. DA-ARO-D-31-124-71-G18.     

Second-order optimality principle for singular optimal control problems

A new necessary condition for singular optimal control problems is presented in this paper. The condition is simpler to apply than existing conditions and is easily derived from a Taylor series expansion of the performance index.