检索结果-内蒙古大学图书馆

11th IFAC symposium on dynamics and Control of Process Systems including Biosystems

作者： Morinelly, Juan E. Ydstie, B. Erik Carnegie Mellon Univ Dept Chem Engn Pittsburgh PA 15213 USA

An adaptive optimal control algorithm for system with uncertain dynamics is formulated under a reinforcement learning framework. An embedded exploratory component, is included explicitly in the objective function of an output feedback receding horizon Model Predictive Control problem. The optimization is formulated as a Quadratically Constrained Quadratic Program and it is solved to epsilon-global optimality. The iterative interaction between the action specified by the optimal solution and the approximation of cost functions balances the exploitation of current knowledge and the need for exploration. The proposed method is shown to converge to the optimal policy for a controllable discrete time linear plant with unknown output parameters. (C) 2016, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.

关键词： adaptive control dual control optimal control model predictive control reinforcement learning approximate dynamic programming

来源：评论

学校读者我要写书评

暂无评论

A New Discrete-Time Iterative adaptive dynamic programming Algorithm Based on Q-learning 12th

引用

12th International symposium on Neural Networks (ISNN)

作者： Wei, Qinglai Liu, Derong Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Univ Sci & Technol Beijing Sch Automat & Elect Engn Beijing 100083 Peoples R China

ISBN: (纸本)9783319253930;9783319253923

In this paper, a novel Q-learning based policy iteration adaptive dynamic programming (ADP) algorithm is developed to solve the optimal control problems for discrete-time nonlinear systems. The idea is to use a policy iteration ADP technique to construct the iterative control law which stabilizes the system and simultaneously minimizes the iterative Q function. Convergence property is analyzed to show that the iterative Q function is monotonically non-increasing and converges to the solution of the optimality equation. Finally, simulation results are presented to show the performance of the developed algorithm.

关键词： adaptive critic designs adaptive dynamic programming approximate dynamic programming Q-learning policy iteration neural networks nonlinear systems optimal control

来源：评论

学校读者我要写书评

暂无评论

A Novel Resilient Control Scheme for a Class of Markovian Jump Systems With Partially Unknown Information

引用

ieee TRANSACTIONS ON CYBERNETICS 2022年第8期52卷 8191-8200页

作者： Zhang, Kun Su, Rong Zhang, Huaguang Chinese Acad Sci Acad Math & Syst Sci Beijing 100190 Peoples R China Nanyang Technol Univ Sch Elect & Elect Engn Singapore 639798 Singapore Northeastern Univ Key Lab Synthet Automat Proc Ind Shenyang 110819 Peoples R China Northeastern Univ Sch Informat Sci & Engn Shenyang 110819 Peoples R China

In the complex practical engineering systems, many interferences and attacking signals are inevitable in industrial applications. This article investigates the reinforcement learning (RL)-based resilient control algorithm for a class of Markovion jump systems with completely unknown transition probability information. Based on the Takagi-Sugeno logical structure, the resilient control problem of the nonlinear Markovion systems is converted into solving a set of local dynamic games, where the control policy and attacking signal are considered as two rival players. Combining the potential learning and forecasting abilities, the new integral RL (IRL) algorithm is designed via system data to compute the zero-sum games without using the information of stationary transition probability. Besides, the matrices of system dynamics can also be partially unknown, and the new architecture requires less transmission and computation during the learning process. The stochastic stability of the system dynamics under the developed overall resilient control is guaranteed based on the Lyapunov theory. Finally, the designed IRL-based resilient control is applied to a typical multimode robot arm system, and implementing results demonstrate the practicality and effectiveness.

关键词： Games Process control Markov processes Game theory Actuators System dynamics Heuristic algorithms adaptive dynamic programming integral reinforcement learning (IRL) resilient control zero-sum game

来源：评论

学校读者我要写书评

暂无评论

adaptive dynamic programming Based Motion Control of Autonomous Underwater Vehicles 5

Adaptive Dynamic Programming Based Motion Control of Autonom...

引用

5th International Conference on Control, Decision and Information Technologies (CoDIT)

作者： Vibhute, Siddhant VJTI Dept Elect Engn Mumbai Maharashtra India

ISBN: (纸本)9781538650653

In this paper, adaptive dynamic programming (ADP) technique is utilized to achieve optimal motion control of Autonomous Underwater Vehicle (AUV) System. The paper proposes a model-free based method that takes into consideration the actuator input and obstacle position while tracing an optimal path. The concept of machine learning enables to develop a path-planner which aims to avoid collisions with static obstacles. The ADP approach is realized to approximate the solution of the cost functional for optimization purpose by which the positions of the locally situated obstacles need not be priori-known until they are within a designed approximation safety envelope. The methodology is implemented to achieve the path-planning objective using dynamic programming technique. The Least-squares policy method serves as a recursive algorithm to approximate the value function for the domain, providing an approach for the finite space discrete control system. The concept behind the design of an obstacle-free path finder is to generate an optimal action that minimizes the local cost, defined by a functional, under constrained optimization. The most advantageous value function is described by the Hamilton Jacobi Bellman (HJB) equation, that is impractical to solve using analytical methods. To overcome the complex calculations subject to HJB, a method based on reinforcement learning (RL), called ADP is implemented. This paper outlines the concept of machine learning to realize a real time obstacle avoidance system.

关键词： Mathematical model dynamic programming Vehicle dynamics Kinematics Optimization Approximation algorithms Control systems

来源：评论

学校读者我要写书评

暂无评论

Advances in reinforcement learning and their implications for intelligent control

Advances in reinforcement learning and their implications fo...

引用

Proceedings of the 5th ieee International symposium on Intelligent Control 1990

作者： Whitehead, Steven D. Sutton, Richard S. Ballard, Dana H. Dept of Comput Sci Univ of Rochester NY USA

ISBN: (纸本)0818621087

The focus of this work is on control architectures that are based on reinforcement learning. A number of recent advances that have contributed to the viability of reinforcement learning approaches to intelligent control are surveyed. These advances include the formalization of the relationship between reinforcement learning and dynamic programming, the use of internal predictive models to improve learning rate, and the integration of reinforcement learning with active perception. On the basis of these advances and other results, it is concluded that control architectures based on reinforcement learning are now in a position to satisfy many of the criteria associated with intelligent control.

关键词： learning Systems

来源：评论

学校读者我要写书评

暂无评论

adaptive dynamic programming-based State Quantized Networked Control System without Value and/or Policy Iterations

Adaptive Dynamic Programming-based State Quantized Networked...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Zhao, Qiming Xu, Hao Jagannathan, S. Missouri Univ S&T Dept Elect & Comp Engn Rolla MO 65409 USA

ISBN: (纸本)9781467314909

In this paper, the Bellman equation is used to solve the stochastic optimal control of unknown linear discrete-time system with communication imperfections including random delays, packet losses and quantization. A dynamic quantizer for the sensor measurements is proposed which essentially provides system states to the controller. To eliminate the effect of the quantization error, the dynamics of the quantization error bound and an update law for tuning its range are derived. Subsequently, by using adaptive dynamic programming technique, the infinite horizon optimal regulation of the uncertain NCS is solved in a forward-in-time manner without using value and/or policy iterations by using Q-function and reinforcement learning. The asymptotic stability of the closed-loop system is verified by standard Lyapunov stability theory. Finally, the effectiveness of the proposed method is verified by simulation results.

关键词： Networked Control System Quantization adaptive dynamic programming Optimal Control

来源：评论

学校读者我要写书评

暂无评论

Automatically customizing a powered knee prosthesis with human in the loop using adaptive dynamic programming

Automatically customizing a powered knee prosthesis with hum...

引用

2017 International symposium on Wearable Robotics and Rehabilitation, WeRob 2017

作者： Wen, Yue Brandt, Andrea Si, Jennie Huang, He Helen NCSU UNC Department of Biomedical Engineering NC State University RaleighNC27695-7115 United States University of North Carolina Chapel HillNC27599 United States Department of Electrical Computer and Energy Engineering Arizona State University TempeAZ85281 United States

ISBN: (纸本)9781538643778

In this study, we validated a human-in-the-loop auto-tuner using machine learning to automatically customize powered knee prosthesis control parameters for an amputee subject in real time. The experimental powered knee prosthesis was controlled by a finite state impedance controller, which had 12 configurable impedance control parameters. Using adaptive dynamic programming with reinforcement learning while one transfemoral amputee subject walked with the powered knee prosthesis, the auto-tuner would interact with the human-prosthesis system and learn to configure the high dimension of control parameters. We tested 4 different initial conditions with one unilateral transfemoral amputee subject. The results showed that the auto-tuner discovered the control parameters that allowed amputee subject to generate normative knee kinematics in 3 out of 4 tuning sessions. For all test sessions, the averaged root-mean-square error of the knee kinematics relative to the normative knee kinematics decreased from 6.6 degrees to 4.6 degrees after tuning procedure. © 2017 ieee.

关键词： adaptive control systems

来源：评论

学校读者我要写书评

暂无评论

Event-Triggered Optimal Neuro-Controller Design With reinforcement learning for Unknown Nonlinear Systems

引用

ieee TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2019年第9期49卷 1866-1878页

作者： Yang, Xiong He, Haibo Liu, Derong Univ Rhode Isl Dept Elect Comp & Biomed Engn Kingston RI 02881 USA Tianjin Univ Sch Elect & Informat Engn Tianjin 300072 Peoples R China Guangdong Univ Technol Sch Automat Guangzhou 510006 Guangdong Peoples R China

This paper develops an optimal control scheme for continuous-time unknown nonlinear systems using the event-triggering mechanism. Different from designing controllers using the time-triggering mechanism, the event-triggered controller is updated only when the system state deviates more than a certain threshold from a prescribed value. To obtain the event-triggered optimal controller, we develop an identifier-critic architecture under the framework of reinforcement learning. The identifier network, composed of a feedforward neural network (FNN), aims to derive the knowledge of unknown system dynamics, and the critic network, constituted of an FNN, intends to derive the event-triggered optimal controller. The identifier network is tuned via the combination of a standard back-propagation algorithm and an e-modification method, and the critic network is updated using a modification of the gradient descent method. By introducing an additional stability term to update the critic network, the initial admissible control is no longer required. Meanwhile, by using historical and instantaneous state data together, the persistence of excitation condition is relaxed. A stability analysis of the closed-loop system is provided based on the Lyapunov method. The effectiveness of the proposed designs is illustrated through simulations of a nonlinear example and a single link robot arm system.

关键词： adaptive dynamic programming (ADP) event-triggered control neural networks (NNs) nonlinear systems optimal control reinforcement learning (RL)

来源：评论

学校读者我要写书评

暂无评论

adaptive Constrained Optimal Control Design for Data-Based Nonlinear Discrete-Time Systems With Critic-Only Structure

引用

ieee TRANSACTIONS ON NEURAL NETWORKS AND learning SYSTEMS 2018年第6期29卷 2099-2111页

作者： Luo, Biao Liu, Derong Wu, Huai-Ning Chinese Acad Sci State Key Lab Management & Control Complex Syst Inst Automat Beijing 100190 Peoples R China Guangdong Univ Technol Sch Automat Guangzhou 510006 Guangdong Peoples R China Beihang Univ Sci & Technol Aircraft Control Lab Beijing 100191 Peoples R China

reinforcement learning has proved to be a powerful tool to solve optimal control problems over the past few years. However, the data-based constrained optimal control problem of nonaffine nonlinear discrete-time systems has rarely been studied yet. To solve this problem, an adaptive optimal control approach is developed by using the value iteration-based Q-learning (VIQL) with the critic-only structure. Most of the existing constrained control methods require the use of a certain performance index and only suit for linear or affine nonlinear systems, which is unreasonable in practice. To overcome this problem, the system transformation is first introduced with the general performance index. Then, the constrained optimal control problem is converted to an unconstrained optimal control problem. By introducing the action-state value function, i.e., Q-function, the VIQL algorithm is proposed to learn the optimal Q-function of the data-based unconstrained optimal control problem. The convergence results of the VIQL algorithm are established with an easy-to-realize initial condition Q((0))(x, a) >= 0. To implement the VIQL algorithm, the critic-only structure is developed, where only one neural network is required to approximate the Q-function. The converged Q-function obtained from the critic-only VIQL method is employed to design the adaptive constrained optimal controller based on the gradient descent scheme. Finally, the effectiveness of the developed adaptive control method is tested on three examples with computer simulation.

关键词： adaptive control adaptive dynamic programming constraints critic-only data-based optimal control Q-learning

来源：评论

学校读者我要写书评

暂无评论

Geometric deep reinforcement learning for dynamic DAG scheduling

Geometric deep reinforcement learning for dynamic DAG schedu...

引用

ieee symposium Series on Computational Intelligence (ieee SSCI)

作者： Grinsztajn, Nathan Beaumont, Olivier Jeannot, Emmanuel Preux, Philippe Univ Lille CNRS UMR 9189 CRIStAL INRIA Lille France Inria Bordeaux Hiepacs Team Bordeaux France Inria Bordeaux TADaaM Team Bordeaux France

ISBN: (纸本)9781728125473

In practice, it is quite common to face combinatorial optimization problems which contain uncertainly along with non determinism and dynamicity. These three properties call for appropriate algorithms;reinforcement learning (RL) is dealing with them in a very natural way. Today, despite some efforts, most real-life combinatorial optimization problems remain out of the reach of reinforcement learning algorithms. In this paper, we propose a reinforcement learning approach to solve a realistic scheduling problem, and apply it to an algorithm commonly executed in the high performance computing community, the CHOLESKY factorization. On the contrary to static scheduling, where tasks are assigned to processors in a predetermined ordering before the beginning of the parallel execution, our method is dynamic: task allocations and their execution ordering are decided at runtime, based on the system state and unexpected events, which allows much more flexibility. To do so, our algorithm uses graph neural networks in combination with an actor critic algorithm (A2C) to build an adaptive representation of the problem on the fly. We show that this approach is competitive with state-of-the-art heuristics used in high performance computing runtime systems. Moreover, our algorithm does not require an explicit model of the environment, but we demonstrate that extra knowledge can easily he incorporated and improves the performance. We also exhibit key properties provided by this RL approach, and study its transfer abilities to other instances.

关键词： reinforcement learning scheduling task graph DAG high performance computing combinatorial optimization

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：