Equivalences between differential games and optimal controls

It is shown that linear differential pursuit games with linear targets, if player controls are required to be piecewise constant or if player controls areL ₁-functions (but pursuer control is bang-bang whenever quarry control is), are equivalent to a linear, autonomous control problem. As a byproduct, a sufficient condition for terminating the game, in Pontryagin's sense, is obtained.The present paper has been influenced by Prof. O. Hajek's work in differential games; the converse parts of the proofs presented here are very similar to those in Ref. 5. The author wishes to thank Dr. Hajek for his suggestions, comments, and critique. The Brasilian Government BNDE provided partial financial support.     

Game of two cars: Case of variable speed

A stochastic version of Isaacs's (Ref. 1) game of two cars is dealt with here. In this version, the pursuer, owing to thrust and drag forces, has a variable speed, whereas the evader's speed is constant. Also, the pursuer can maneuver as long as his speed is bounded by some lower and upper limits. The probability of interception, corresponding to optimal (saddle-point) feedback strategies, is computed and serves as a reference for evaluating the performance of four different versions of the proportional navigation pursuit law as well as two other strategies.     

二级线性价格控制问题的满意解的求解思路及实例

本文就文献 [1]对线性二级价格控制问题 (BLP) 2 研究的结果及文献 [2 ]提出的问题进行了进一步的讨论。用反例指出文 [1]求出的极点最优解是错的 ,以及一般的 (BL P) 2 问题求出的最优解 ,可能是下层决策者根本无法接受的。因此 ,本文提出 (BLP) 2 的求下层目标最优解的边界搜索法及在此基础上用多目标的观点来求 (BLP) 2 的满意解的思路及实例     

Expected number of steps of a random optimization method

Some explanatory remarks are presented, relevant to the objections by Pang (Ref. 1) to Dorea's paper (Ref. 2).     

On a student-related optimal control problem

In a recent paper (Ref. 1), Cheng and Teo discussed some further extensions of a student-related optimal control problem which was originally proposed by Raggettet al. (Ref. 2) and later on modified by Parlar (Ref. 3). In this paper, we treat further extensions of the problem.This paper is a modified and improved version of Ref. 4. It is based, in part, on research sponsored by NSF.     

高速扩展平面应力裂纹尖端的各向异性塑性场的补充研究(Ⅰ)

文献[1]的结果对于β≥2的情形不适用,为此,我们用文献[1]和[2]的方法导出了β=2和β>2两者的高速扩展平面应力裂纹尖端的各向异性塑性场的一般表达式.     

Minimax Results and Finite-Dimensional Separation

In this paper, we review and unify some classes of generalized convex functions introduced by different authors to prove minimax results in infinite-dimensional spaces and show the relations between these classes. We list also for the most general class already introduced by Jeyakumar (Ref. 1) an elementary proof of a minimax result. The proof of this result uses only a finite-dimensional separa- tion theorem; although this minimax result was already presented by Neumann (Ref. 2) and independently by Jeyakumar (Ref. 1), we believe that the present proof is shorter and more transparent.     

On the Optimum Radius of Robust Stability for Schur Polynomials

A recent paper (Ref. 1) established a new approach to estimate the robust stability radius of a Schur polynomial. This note points out that the approach given in Ref. 1 is not correct and also gives a counterexample to the main result of Ref. 1.     

Technical Note Discontinuous Implicit Quasivariational Inequalities in Normed Spaces

In this paper, we consider an implicit quasivariational inequality without continuity assumptions in normed spaces. The main result (Theorem 2.1) provides an infinite-dimensional version of Theorem 3.2 in Ref. 1. To achieve such a goal, we employ Theorem 3.2 in Ref. 1 and the technique of Cubiotti in Ref. 2. In particular, Theorem 3.1 covers a recent result of Cubiotti (Theorem 3.1 of Ref. 2) as a special case. Communicated by F. Giannessi This research was partially supported by the National Science Council of Taiwan, ROC.     

Game-Theoretic Approach to the Optimization of FDDI Computer Networks1

The problem of maximizing the performance in a fiber distributed data interface (FDDI) computer network is formulated as a cooperative n-player game. Solutions of this game can be obtained by solving special optimization problems. Using the models and formulas developed by Tangemann (Ref. 1) and Klehmet (Ref. 2) for the mean waiting times, the resulting optimization problems are presented and numerical results are given.     

On the Probability of Existence of Pure Equilibria in Matrix Games

In a recent paper (Ref. 1), Papavassilopoulos obtained results on the probability of the existence of pure equilibrium solutions in stochastic matrix games. We report a similar result, but where the payoffs are drawn from a finite set of numbers N. In the limiting case, as N tends to infinity, our result and that of Papavassilopoulos are identical. We also cite similar results obtained independently by others, some of which were already independently brought to the notice of Papavassilopoulos by Li Calzi as reported in Papavassilopoulos (Ref. 2). We cite a much earlier result obtained by Goldman (Ref. 3). We also cite our related work (Ref. 4), in which we derive the conditions for the existence of mixed strategy equilibria in two-person zero-sum games.     

Remarks on bang-bang control in Hilbert space

In this note, a natural definition of bang-bang control in Hilbert space is given, and some of the theory of the authors' paper (Ref. 1) is rebuilt upon it. An elliptic boundary-value problem illustrating the theory is given. In the last part of this note, the results of Ref. 1 are extended to nonlinear perturbations of linear operators and to homogeneous nonlinear operators.Communicated by L. CesariThis research was supported by NRC Grant No. A-4047 and NSF Grant No. GP-7445.     

An Extension of the Coordinate Transformation Method for Open-Loop Nash Equilibria

In Leitmann (Ref. 1), a coordinate transformation method was introduced to obtain global solutions for free problems in the calculus of variations. This direct method was extended and broadened in Carlson (Ref. 2) and later in Leitmann (Ref. 3). The applicability of the original work of Leitmann (Ref. 1) was further developed in Dockner and Leitmann (Ref. 4) to include the class of open-loop dynamic games. In the present work, we improve the results of Ref. 4 in two directions. First, we enlarge the class of open-loop dynamic games to permit coupling among the dynamic equations via the states of the players; second, we incorporate the modifications given in Refs. 2 and 3. Our results greatly increase the applicability of this method. An example arising from the harvesting of a renewable resource is presented to illustrate the utility of our results.     

On convex vectorial optimization in linear spaces

We observe that the results and proofs of Ref. 1 are valid only for finite convex functionals, and we give some other corrections to Ref. 1.     

Software Release Games

The software release game developed in Ref. 1 is reconsidered in the framework of a two-person nonzero-sum game of timing. More precisely, noisy-type software release strategies are derived in closed form under two different criteria as well as an alternative silent-type software release strategy. Our method overcomes the fatal problem in Ref. 1 and has an advantage on computational tractability. Also, the method can be extended directly to obtain noisy-type strategies.     

Least-Square Fitting with Spheres

Fitting circles and spheres to given data in is at least relevant in computational metrology (Ref. 1) and reflectrometry (Ref. 2). A new descent algorithm, developed for circles in Ref. 3, is generalized to spheres. Numerical examples are given.     

On the Tikhonov Well-Posedness of Concave Games and Cournot Oligopoly Games

The purpose of this paper is to investigate whether theorems known to guarantee the existence and uniqueness of Nash equilibria, provide also sufficient conditions for the Tikhonov well-posedness (T-wp). We consider several hypotheses that ensure the existence and uniqueness of a Nash equilibrium (NE), such as strong positivity of the Jacobian of the utility function derivatives (Ref. 1), pseudoconcavity, and strict diagonal dominance of the Jacobian of the best reply functions in implicit form (Ref. 2). The aforesaid assumptions imply the existence and uniqueness of NE. We show that the hypotheses in Ref. 2 guarantee also the T-wp property of the Nash equilibrium.As far as the hypotheses in Ref. 1 are concerned, the result is true for quadratic games and zero-sum games. A standard way to prove the T-wp property is to show that the sets of -equilibria are compact. This last approach is used to demonstrate directly the T-wp property for the Cournot oligopoly model given in Ref. 3. The compactness of -equilibria is related also to the condition that the best reply surfaces do not approach each other near infinity.     

Intersection Theorems with a Continuum of Intersection Points

In all existing intersection theorems, conditions are given under which a certain subset of a collection of sets has a nonempty intersection. In this paper, conditions are formulated under which the intersection is a continuum of points satisfying some interesting topological properties. In this sense, the intersection theorems considered in this paper belong to a new class. The intersection theorems are formulated on the unit cube and it is shown that both the vector of zeroes and the vector of ones lie in the same component of the intersection. An interesting application concerns the model of an economy with price rigidities. Using the intersection theorems of this paper, it is easily shown that there exists a continuum of zero points in such a model. The intersection theorems treated give a generalization of the well-known lemmas of Knaster, Kuratowski, and Mazurkiewicz (Ref. 1), Scarf (Ref. 2), Shapley (Ref. 3), and Ichiishi (Ref. 4). Moreover, the results can be used to sharpen the usual formulation of the Scarf lemma on the cube.     

Decomposition of the convex simplex method

Rutenberg (Ref. 1) provided a decomposition method to solve the problem of minimizing a separable, nonlinear objective function with large-scale linear constraints by using the convex simplex method (Ref. 2). However, there seem to be some errors in his paper (Ref. 3). This paper will rebuild the method by using more convenient and consistent notation.     

A strong variational algorithm for delay systems

In this paper, we present a convergent extension of the first-order strong-variational algorithm by Mayne and Polak (Ref. 1) for solving optimal control problems with control constraints to delay systems. Although the algorithm is similar to the one presented in Ref. 1, the proof of convergence is different, since the differential dynamic techniques used by Mayne and Polak are not applicable.This work forms part of the author's PhD Dissertation and was conducted at the Imperial College of Science and Technology under a studentship awarded by the UK Science and Engineering Research Council. This assistance is gratefully acknowledged. The author also wishes to thank Dr. R. B. Vinter for his encouragement and help.