Sufficient optimality conditions for nonconvex control problems with state constraints

The sufficient optimality conditions of Zeidan for optimal control problems (Refs. 1 and 2) are generalized such that they are applicable to problems with pure state-variable inequality constraints. We derive conditions which neither assume the concavity of the Hamiltonian nor the quasiconcavity of the constraints. Global as well as local optimality conditions are presented.     

Necessary conditions for optimal control problems with state constraints

Necessary conditions of optimality are derived for optimal control problems with pathwise state constraints, in which the dynamic constraint is modelled as a differential inclusion. The novel feature of the conditions is the unrestrictive nature of the hypotheses under which these conditions are shown to be valid. An Euler Lagrange type condition is obtained for problems where the multifunction associated with the dynamic constraint has values possibly unbounded, nonconvex sets and satisfies a mild `one-sided' Lipschitz continuity hypothesis. We recover as a special case the sharpest available necessary conditions for state constraint free problems proved in a recent paper by Ioffe. For problems where the multifunction is convex valued it is shown that the necessary conditions are still valid when the one-sided Lipschitz hypothesis is replaced by a milder, local hypothesis. A recent `dualization' theorem permits us to infer a strengthened form of the Hamiltonian inclusion from the Euler Lagrange condition. The necessary conditions for state constrained problems with convex valued multifunctions are derived under hypotheses on the dynamics which are significantly weaker than those invoked by Loewen and Rockafellar to achieve related necessary conditions for state constrained problems, and improve on available results in certain respects even when specialized to the state constraint free case.

Proofs make use of recent `decoupling' ideas of the authors, which reduce the optimization problem to one to which Pontryagin's maximum principle is applicable, and a refined penalization technique to deal with the dynamic constraint.

     

Sufficient conditions for optimality in problems with state constraints

In this paper the sufficient conditions for optimality are obtained for problems with state constraints. These constraints may be active. It means that the adjoint function may have points of discontinuity or jumps. Similar results in the case of absolutely continuous adjoint function were given by the author in [1] and [2].     

Sufficient conditions for asymptotic optimal control

For a selected family of Lagrange-type control problems involving a nonnegative integral costJ _T (y,u) over the interval [0,T], 0<T<, with system conditions consisting of differential inequalities and/or equalities, the following material is treated: (i) a resumé of relevant necessary conditions and sufficient conditions for a pair (y _T ,u _T ) to minimizeJ _T (y,u); (ii) conditions sufficient for the convergence asT of minimizing pairs (y _T ,u _T ) over [0,T] to a limit pair (y ,u ) over the infinite-time interval [0, ); (iii) conditions sufficient for (y ,u ) to minimize the costJ (y,u) over [0, ); and (iv) conditions sufficient for the optimal cost per unit timeJ _T (y _T ,u _T )/T to have a limit asT.     

Arrow-type sufficient optimality conditions for nondifferentiable optimal control problems with state constraints

In this paper a class of nondifferentiable optimal control problems governed by differential inclusions and subject to state variable inequality constraints is considered. Sufficient conditions using the concavity of the maximized Hamiltonian are given. Furthermore, a counterexample is presented that shows that in the nondifferentiable case the maximum principle does not form sufficient optimality conditions if the adjoint relation is formulated in terms of the ordinary Hamiltonian rather than the maximized one. Finally, it is shown that the sufficient conditions correspond to Clarke's necessary conditions with some additional assumptions such as concavity.     

Nondegenerate necessary conditions for nonlinear optimal control problems with higher-index state constraints

For some optimal control problems with pathwise state constraints the standard versions of the necessary conditions of optimality are unable to provide useful information to select minimizers. There exist some literature on stronger forms of the maximum principle, the so-called nondegenerate necessary conditions, that can be informative for those problems. These conditions can be applied when certain constraint qualifications are satisfied. However, when the state constraints have higher index (i.e. their first derivative with respect to time does not depend on the control) these nondegenerate necessary conditions cannot be used. This happens because constraint qualifications assumptions are never satisfied for higher index state constraints. We note that control problems with higher index state constraints arise frequently in practice. An example is a common mechanical systems for which there is a constraint on the position (an obstacle in the path, for example) and the control acts as a second derivative of the position (a force or acceleration) which is a typical case. Here, we provide a nondegenerate form of the necessary conditions that can be applied to nonlinear problems with higher index state constraints. When addressing a problem with a state constraint of index k, the result described is applicable under a constraint qualification that involves the k -th derivative of the state constraint, corresponding to the first time when derivative depends explicitly on the control. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Sufficient quadratic conditions of extremum for discontinuous controls in optimal control problems with mixed constraints

We derive second-order sufficient optimality conditions for discontinuous controls in optimal control problems of ordinary differential equations with initial-final state constraints and mixed state-control constraints of equality and inequality type. Under the assumption that the gradients with respect to the control of active mixed constraints are linearly independent, the sufficient conditions imply a bounded strong minimum in the problem.     

Necessary extremum conditions in the optimal control problem with state constraints

The optimal control problem with state constraints is examined. An alternative to the available approaches to the study of this problem is proposed. The maximum principle and second-order necessary conditions are proved.     

Sufficient conditions for optimal control problems with time delay

Various first-order and second-order sufficient conditions of optimality for nonlinear optimal control problems with delayed argument are formulated. The functions involved are not required to be convex. Second-order sufficient conditions are shown to be related to the existence of solutions of a Riccati-type matrix differential inequality. Their relation with the second variation is discussed.The authors are indebted to an anonymous referee for valuable suggestions that lead to various improvements in the paper.     

Sufficient optimality conditions for multidimensional optimal control problems with mixed constraints governed by an elliptic PDE

In this article, the idea of a dual dynamic programming is applied to the optimal control problems with multiple integrals governed by a semi-linear elliptic PDE and mixed state-control constraints. The main result called a verification theorem provides the new sufficient conditions for optimality in terms of a solution to the dual equation of a multidimensional dynamic programming. The optimality conditions are also obtained by using the concept of an optimal dual feedback control. Besides seeking the exact minimizers of problems considered some kind of an approximation is given and the sufficient conditions for an approximated optimal pair are derived.     

Sufficient second-order optimality conditions for convex control constraints

In this article sufficient optimality conditions are established for optimal control problems with pointwise convex control constraints. Here, the control is a function with values in Rn. The constraint is of the form u(x)∈U(x), where U is a set-valued mapping that is assumed to be measurable with convex and closed images. The second-order condition requires coercivity of the Lagrange function on a suitable subspace, which excludes strongly active constraints, together with first-order necessary conditions. It ensures local optimality of a reference function in an L^∞-neighborhood. The analysis is done for a model problem namely the optimal distributed control of the instationary Navier-Stokes equations.     

On optimal control problems with state constraints

This paper is concerned with necessary conditions for a general optimal control problem developed by Russak and Tan. It is shown that, in most cases, a further relation between the multipliers holds. This result is of interest in particular for the investigation of perturbations of the state constraint.     

Boundary optimal flow control with state constraints

The numerical solution of the Dirichlet boundary optimal control problem of the Navier-Stokes equations in presence of pointwise state constraints is investigated. A Moreau-Yosida regularization of the problem is proposed to obtain regular multipliers. Optimality conditions are derived and the convergence of the regularized solutions towards the original one is presented. The paper ends with a numerical experiment. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Error estimates for parabolic optimal control problems with control and state constraints

The numerical approximation to a parabolic control problem with control and state constraints is studied in this paper. We use standard piecewise linear and continuous finite elements for the space discretization of the state, while the dG(0) method is used for time discretization. A priori error estimates for control and state are obtained by an improved maximum error estimate for the corresponding discretized state equation. Numerical experiments are provided which support our theoretical results.     

A-posteriori error estimates for optimal control problems with state and control constraints

We discuss the full discretization of an elliptic optimal control problem with pointwise control and state constraints. We provide the first reliable a-posteriori error estimator that contains only computable quantities for this class of problems. Moreover, we show, that the error estimator converges to zero if one has convergence of the discrete solutions to the solution of the original problem. The theory is illustrated by numerical tests.     

Approximation of optimal control problems with state constraints

We study the approximation of control problems governed by elliptic partial differential equations with pointwise state constraints. For a finite dimensional approximation of the control set and for suitable perturbations of the state constraints, we prove that the corresponding sequence of discrete control problems converges to a relaxed problem. A similar analysis is carried out for problems in which the state equation is discretized by a finite element method.     

Sufficient conditions for optimality in two-person zero-sum differential games with state and strategy constraints

A sufficiency theorem for optimal feedback strategies in two-person zero-sum differential games is given. The theorem is applicable to a wide class of such games for which strategies are Borel measurable functions on a subset of the state space. The theorem generalizes those of [1, 2, and 5].     

Quadratic weak-minimum conditions for optimal control problems with intermediate constraints

Optimal control problems with constraints at intermediate trajectory points are considered. By using a certain natural method (of reproduction of state and control variables), these problems reduce to the standard optimal control problem of Pontryagin type, which allows one to obtain quadratic weak-minimum conditions for them.     

Second-order analysis for optimal control problems with pure state constraints and mixed control-state constraints

This paper deals with the optimal control problem of an ordinary differential equation with several pure state constraints, of arbitrary orders, as well as mixed control-state constraints. We assume (i) the control to be continuous and the strengthened Legendre–Clebsch condition to hold, and (ii) a linear independence condition of the active constraints at their respective order to hold. We give a complete analysis of the smoothness and junction conditions of the control and of the constraints multipliers. This allows us to obtain, when there are finitely many nontangential junction points, a theory of no-gap second-order optimality conditions and a characterization of the well-posedness of the shooting algorithm. These results generalize those obtained in the case of a scalar-valued state constraint and a scalar-valued control.