عنوان

Reinforcement learning for optimal feedback control :

(Number (ISBN

331978384X

(Number (ISBN

9783319783840

Erroneous ISBN

3319783831

Erroneous ISBN

9783319783833

Title Proper

Reinforcement learning for optimal feedback control :

General Material Designation

[Book]

Other Title Information

a Lyapunov-based approach /

First Statement of Responsibility

Rushikesh Kamalapurkar, Patrick Walters, Joel Rosenfeld, Warren Dixon.

Place of Publication, Distribution, etc.

Cham, Switzerland :

Name of Publisher, Distributor, etc.

Springer,

Date of Publication, Distribution, etc.

2018.

Specific Material Designation and Extent of Item

1 online resource (xvi, 293 pages) :

Other Physical Details

illustrations

Series Title

Communications and control engineering

Text of Note

Includes bibliographical references and index.

Text of Note

Intro; Preface; Contents; Symbols; 1 Optimal Control; 1.1 Introduction; 1.2 Notation; 1.3 The Bolza Problem; 1.4 Dynamic Programming; 1.4.1 Necessary Conditions for Optimality; 1.4.2 Sufficient Conditions for Optimality; 1.5 The Unconstrained Affine-Quadratic Regulator; 1.6 Input Constraints; 1.7 Connections with Pontryagin's Maximum Principle; 1.8 Further Reading; 1.8.1 Numerical Methods; 1.8.2 Differential Games and Equilibrium Solutions; 1.8.3 Viscosity Solutions and State Constraints; References; 2 Approximate Dynamic Programming; 2.1 Introduction

Text of Note

2.2 Exact Dynamic Programming in Continuous Time and Space; 2.2.1 Exact Policy Iteration: Differential and Integral Methods; 2.2.2 Value Iteration and Associated Challenges; 2.3 Approximate Dynamic Programming in Continuous Time and Space; 2.3.1 Some Remarks on Function Approximation; 2.3.2 Approximate Policy Iteration; 2.3.3 Development of Actor-Critic Methods; 2.3.4 Actor-Critic Methods in Continuous Time and Space; 2.4 Optimal Control and Lyapunov Stability; 2.5 Differential Online Approximate Optimal Control; 2.5.1 Reinforcement Learning-Based Online Implementation

Text of Note

2.5.2 Linear-in-the-Parameters Approximation of the Value Function; 2.6 Uncertainties in System Dynamics; 2.7 Persistence of Excitation and Parameter Convergence; 2.8 Further Reading and Historical Remarks; References; 3 Excitation-Based Online Approximate Optimal Control; 3.1 Introduction; 3.2 Online Optimal Regulation; 3.2.1 Identifier Design; 3.2.2 Least-Squares Update for the Critic; 3.2.3 Gradient Update for the Actor; 3.2.4 Convergence and Stability Analysis; 3.2.5 Simulation; 3.3 Extension to Trajectory Tracking; 3.3.1 Formulation of a Time-Invariant Optimal Control Problem

Text of Note

3.3.2 Approximate Optimal Solution; 3.3.3 Stability Analysis; 3.3.4 Simulation; 3.4 N-Player Nonzero-Sum Differential Games; 3.4.1 Problem Formulation; 3.4.2 Hamilton-Jacobi Approximation Via Actor-Critic-Identifier; 3.4.3 System Identifier; 3.4.4 Actor-Critic Design; 3.4.5 Stability Analysis; 3.4.6 Simulations; 3.5 Background and Further Reading; References; 4 Model-Based Reinforcement Learning for Approximate Optimal Control; 4.1 Introduction; 4.2 Model-Based Reinforcement Learning; 4.3 Online Approximate Regulation; 4.3.1 System Identification; 4.3.2 Value Function Approximation

Text of Note

4.3.3 Simulation of Experience Via Bellman Error Extrapolation; 4.3.4 Stability Analysis; 4.3.5 Simulation; 4.4 Extension to Trajectory Tracking; 4.4.1 Problem Formulation and Exact Solution; 4.4.2 Bellman Error; 4.4.3 System Identification; 4.4.4 Value Function Approximation; 4.4.5 Simulation of Experience; 4.4.6 Stability Analysis; 4.4.7 Simulation; 4.5 N-Player Nonzero-Sum Differential Games; 4.5.1 System Identification; 4.5.2 Model-Based Reinforcement Learning; 4.5.3 Stability Analysis; 4.5.4 Simulation; 4.6 Background and Further Reading; References; 5 Differential Graphical Games

0

8

8

8

8

Text of Note

Reinforcement Learning for Optimal Feedback Control develops model-based and data-driven reinforcement learning methods for solving optimal control problems in nonlinear deterministic dynamical systems. In order to achieve learning under uncertainty, data-driven methods for identifying system models in real-time are also developed. The book illustrates the advantages gained from the use of a model and the use of previous experience in the form of recorded data through simulations and experiments. The book's focus on deterministic systems allows for an in-depth Lyapunov-based analysis of the performance of the methods described during the learning phase and during execution. To yield an approximate optimal controller, the authors focus on theories and methods that fall under the umbrella of actor-critic methods for machine learning. They concentrate on establishing stability during the learning phase and the execution phase, and adaptive model-based and data-driven reinforcement learning, to assist readers in the learning process, which typically relies on instantaneous input-output measurements. This monograph provides academic researchers with backgrounds in diverse disciplines from aerospace engineering to computer science, who are interested in optimal reinforcement learning functional analysis and functional approximation theory, with a good introduction to the use of model-based methods. The thorough treatment of an advanced treatment to control will also interest practitioners working in the chemical-process and power-supply industry.

Source for Acquisition/Subscription Address

Springer Nature

Stock Number

com.springer.onix.9783319783840

Title

Reinforcement learning for optimal feedback control.

International Standard Book Number

9783319783833

Feedback control systems.

Reinforcement learning.

Automatic control engineering.

Calculus of variations.

Communications engineering-- telecommunications.

COMPUTERS-- General.

Cybernetics & systems theory.

Feedback control systems.

Reinforcement learning.

COM-- 000000

TJFM

Number

Edition

Class number

Dixon, Warren E.,1972-

Kamalapurkar, Rushikesh

Rosenfeld, Joel A.

Walters, Patrick S.

Date of Transaction

20200823113256.0

Cataloguing Rules (Descriptive Conventions))

pn

Electronic name

[Book]

Y