Markov Chain Approximations for Deter- Ministic Control Problems with Aane Dynamics and Quadratic Cost In

نویسندگان

  • W H Fleming
  • H M Soner
  • P E Souganidis
  • M Bou
  • P Dupuis
چکیده

An optimal control formulation and related numerical methods for a problem in shape reconstruction, Annals of Applied Probability, 4 (1994), pp. 287{346. 36 surface z(x; y) = (1=4) sin(7 4 x) sin(7 4 y): The approximate controls are computed on a grid with spacing h = 0:025, and the trajectories are integrated by a simple Euler method with linear interpolation. In Figure 7b, we show the value function computed with h = 0:025 for the corresponding control problem, illustrating the fairly complex structure of the regions of strong regularity. Since we do not know the exact solution for this problem, it is not possible for us to present a numerical measure of accuracy for the approximate geodesics in Figure 7a. However, Theorem 5.5 guarantees that for initial conditions in a region of strong regularity, the approximations will converge to the correct geodesics as h ! 0. Being that a more reened grid does not result in discernible changes to the paths indicated in Figure 7a, we conclude that these are, in fact, good approximations to the exact geodesic curves.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Numerical Approximations for Stochastic Systems With Delays in the State and Control

The Markov chain approximation numerical methods are widely used to compute optimal value functions and controls for stochastic as well as deterministic systems. We extend them to controlled general nonlinear delayed reflected diffusion models. The path, control, and reflection terms can all be delayed. Previous work developed numerical approximations and convergence theorems. But when the cont...

متن کامل

Numerical Approximations for Stochastic Differential Games: The Ergodic Case

The Markov chain approximation method is a widely used, relatively easy to use, and efficient family of methods for the bulk of stochastic control problems in continuous time, for reßected-jump-diffusion type models. It has been shown to converge under broad conditions, and there are good algorithms for solving the numerical problems, if the dimension is not too high. We consider a class of sto...

متن کامل

Learning In

Intelligent systems require software incorporating probabilistic reasoning, and often times learning. Networks provide a framework and methodology for creating this kind of software. This paper introduces network models based on chain graphs with deter-ministic nodes. Chain graphs are deened as a hierarchical combination of Bayesian and Markov networks. To model learning, plates on chain graphs...

متن کامل

Using Markov Chain to Analyze Production Lines Systems with Layout Constraints

There are some problems with estimating the time required for the manufacturing process of products, especially when there is a variable serving time, like control stage. These problems will cause overestimation of process time. Layout constraints, reworking constraints and inflexible product schedule in multi product lines need a precise planning to reduce volume in particular situation of lin...

متن کامل

Multi-point Gaussian states, quadratic-exponential cost functionals, and large deviations estimates for linear quantum stochastic systems

This paper is concerned with risk-sensitive performance analysis for linear quantum stochastic systems interacting with external bosonic fields. We consider a cost functional in the form of the exponential moment of the integral of a quadratic polynomial of the system variables over a bounded time interval. An integro-differential equation is obtained for the time evolution of this quadratic-ex...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1998