检索结果-内蒙古大学图书馆

Deterministic policy gradient adaptive dynamic programming for model-free optimal control

NEUROCOMPUTING 2020年第0期387卷 40-50页

作者： Zhang, Yongwei Zhao, Bo Liu, Derong Guangdong Univ Technol Sch Automat Guangzhou 510006 Peoples R China Beijing Normal Univ Sch Syst Sci Beijing 100875 Peoples R China Northeastern Univ State Key Lab Synthet Automat Proc Ind Shenyang 110819 Peoples R China

In this paper, a deterministic policy gradient adaptive dynamic programming (DPGADP) algorithm is proposed for solving model-free optimal control problems of discrete-time nonlinear systems. By using the measured data, the developed algorithm improves the control performance with the policy gradient method. The convergence of DPGADP algorithm is demonstrated by showing that the constructed Q-function sequence is monotonically non-increasing and converges to the optimal Q-function. An actorcritic neural network (NN) structure is established to implement the DPGADP algorithm. Experience replay and target network techniques from deep Q-learning are employed during the training process. The stability of the NN weight error dynamics is also analyzed. Finally, two simulation examples are presented to verify the effectiveness of the proposed method. (C) 2019 Elsevier B.V. All rights reserved.

关键词： adaptive dynamic programming Deterministic policy gradient Optimal control Neural networks Model-free

来源：评论

学校读者我要写书评

暂无评论

Event-triggered adaptive dynamic programming for discrete-time multi-player games

引用

INFORMATION SCIENCES 2020年 506卷 457-470页

作者： Wang, Ziyang Wei, Qinglai Liu, Derong Univ Sci & Technol Beijing Sch Automat & Elect Engn Beijing 100083 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Guangdong Univ Technol Sch Automat Guangzhou 510006 Guangdong Peoples R China

For multi-player games, the event-triggered adaptive dynamic programming (ADP) method with single triggering condition cannot be used. In this paper, an event-triggered ADP method with multiple triggering conditions is developed for multi-player non-zero-sum (NZS) games. Triggering conditions are designed for each player, and the control inputs will be updated only when the relevant conditions are satisfied. Besides, the developed method is implemented by single-network structure. Therefore, the computational burden can be reduced effectively. Additionally, the stability is analyzed for event-triggered multi-player systems. Finally, two examples are employed to show the effectiveness of the developed method. (C) 2019 Published by Elsevier Inc.

关键词： adaptive dynamic programming Neural network Event-triggered control Multi-player game

来源：评论

学校读者我要写书评

暂无评论

Decentralized robust optimal control for modular robot manipulators via critic-identifier structure-based adaptive dynamic programming

引用

NEURAL COMPUTING & APPLICATIONS 2020年第8期32卷 3441-3458页

作者： Dong, Bo Zhou, Fan Liu, Keping Li, Yuanchun Changchun Univ Technol Dept Control Sci & Engn Changchun 130012 Peoples R China Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China

This paper presents a decentralized robust optimal control method for modular robot manipulators (MRMs) via a critic-identifier structure-based adaptive dynamic programming (ADP) scheme. The robust control problem of MRMs is transformed into an optimal compensation control issue, which consists of model-based compensation control, identifier-based learning control and ADP-based optimal control. The dynamic model of MRMs is deployed for each joint module where the local dynamic information is utilized to design the model compensation controller. A neural network (NN) identifier is established to approximate the interconnected dynamic coupling. Based on ADP and local online policy iteration algorithm, the Hamiltonian-Jacobi-Bellman equation is solved by constructing a critic NN, and then the approximate optimal control policy derivation is possible. The closed-loop robotic system is asymptotic stable by the implementation of a set of developed decentralized control policies. Simulations are presented to demonstrate the effectiveness of the proposed method.

关键词： Modular robot manipulator Decentralized control adaptive dynamic programming Optimal control Neural network Interconnected dynamic coupling

来源：评论

学校读者我要写书评

暂无评论

Invariant adaptive dynamic programming for Discrete-Time Optimal Control

引用

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2020年第11期50卷 3959-3971页

作者： Zhu, Yuanheng Zhao, Dongbin He, Haibo Chinese Acad Sci Inst Automat State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing 100049 Peoples R China Univ Rhode Isl Dept Elect Comp & Biomed Engn Kingston RI 02881 USA

For systems that can only be locally stabilized, control laws and their effective regions are both important. In this paper, invariant policy iteration is proposed to solve the optimal control of discrete-time systems. At each iteration, a given policy is evaluated in its invariantly admissible region, and a new policy and a new region are updated for the next iteration. Theoretical analysis shows the method is regionally convergent to the optimal value and the optimal policy. Combined with sum-of-squares polynomials, the method is able to achieve the near-optimal control of a class of discrete-time systems. An invariant adaptive dynamic programming algorithm is developed to extend the method to scenarios where system dynamics is not available. Online data are utilized to learn the near-optimal policy and the invariantly admissible region. Simulated experiments verify the effectiveness of our method.

关键词： Optimal control Discrete-time systems Heuristic algorithms dynamic programming Convergence Artificial intelligence Nonlinear systems adaptive dynamic programming discrete-time systems invariant admissibility optimal control policy iteration sum of squares

来源：评论

学校读者我要写书评

暂无评论

adaptive dynamic programming-Based Fault-Tolerant Position-Force Control of Constrained Reconfigurable Manipulators

引用

IEEE ACCESS 2020年 8卷 183286-183299页

作者： Ma, Bing Dong, Bo Zhou, Fan Li, Yuanchun Changchun Univ Technol Dept Control Sci & Engn Changchun 130021 Peoples R China

This article presents a novel fault-tolerant position-force optimal control method for constrained reconfigurable manipulators with uncertain actuator failures. On the basis of the radial basis function neural network (RBFNN)-estimated manipulators dynamics, the proposed force-position error fusion function and the estimated actuator failure are utilized to construct an improved optimal performance index function, which reflects the faults and optimizes system comprehensive performance as well as the energy consumption simultaneously. Based on the policy iteration (PI) scheme and the adaptive dynamic programming (ADP) algorithm, the Hamiltonian-Jacobi-Bellman (HJB) equation is solved by constructing the critic neural network (NN), and then the approximated fault-tolerant position-force optimal control policy can be derived correspondingly. The closed-loop manipulator system is proved to be asymptotically stable by using the Lyapunov theory. Finally, simulations are provided to demonstrate the effectiveness of the proposed method.

关键词： Manipulator dynamics Fault tolerance Fault tolerant systems Actuators Optimal control Reconfigurable manipulators adaptive dynamic programming fault-tolerant position-force control optimal control neural network

来源：评论

学校读者我要写书评

暂无评论

adaptive dynamic programming-based controller with admittance adaptation for robot-environment interaction

引用

INTERNATIONAL JOURNAL OF ADVANCED ROBOTIC SYSTEMS 2020年第3期17卷

作者： Zhan, Hong Huang, Dianye Chen, Zhaopeng Wang, Min Yang, Chenguang South China Univ Technol Key Lab Autonomous Syst & Networked Control Minist Educ Sch Automat Sci & Engn Guangzhou Peoples R China Univ Hamburg Dept Informat TAMS Grp D-22527 Hamburg Germany Univ West England Bristol Robot Lab Bristol BS16 1QY Avon England

The problem of optimal tracking control for robot-environment interaction is studied in this article. The environment is regarded as a linear system and an admittance control with iterative linear quadratic regulator method is obtained to guarantee the compliant behaviour. Meanwhile, an adaptive dynamic programming-based controller is proposed. Under adaptive dynamic programming frame, the critic network is performed with radial basis function neural network to approximate the optimal cost, and the neural network weight updating law is incorporated with an additional stabilizing term to eliminate the requirement for the initial admissible control. The stability of the system is proved by Lyapunov theorem. The simulation results demonstrate the effectiveness of the proposed control scheme.

关键词： adaptive dynamic programming admittance adaptation neural network optimal control robot-environment interaction

来源：评论

学校读者我要写书评

暂无评论

The adaptive dynamic programming Toolbox

引用

SENSORS 2021年第16期21卷 5609-5609页

作者： Xing, Xiaowei Chang, Dong Eui Korea Adv Inst Sci & Technol Sch Elect Engn Daejeon 34141 South Korea

The paper develops the adaptive dynamic programming toolbox (ADPT), which is a MATLAB-based software package and computationally solves optimal control problems for continuous-time control-affine systems. The ADPT produces approximate optimal feedback controls by employing the adaptive dynamic programming technique and solving the Hamilton-Jacobi-Bellman equation approximately. A novel implementation method is derived to optimize the memory consumption by the ADPT throughout its execution. The ADPT supports two working modes: model-based mode and model-free mode. In the former mode, the ADPT computes optimal feedback controls provided the system dynamics. In the latter mode, optimal feedback controls are generated from the measurements of system trajectories, without the requirement of knowledge of the system model. Multiple setting options are provided in the ADPT, such that various customized circumstances can be accommodated. Compared to other popular software toolboxes for optimal control, the ADPT features computational precision and time efficiency, which is illustrated with its applications to a highly non-linear satellite attitude control problem.

关键词： adaptive dynamic programming optimal control software package

来源：评论

学校读者我要写书评

暂无评论

Thermoelectric Tracking Control of Solid Oxide Fuel Cell System Using adaptive dynamic programming

Thermoelectric Tracking Control of Solid Oxide Fuel Cell Sys...

引用

第34届中国控制与决策会议

作者： Xiao-long Wu Hongli Liu Hong Zhang Jinting Fu Bin Xie Han Hu Xi Li School of information engineering Nanchang University School of Artificial and Automation Key Laboratory of Education Ministry for Image Processing and Intelligent ControlHuazhong University of Science & Technology

An adaptive optimal control strategy for solid oxide fuel cell power generation systems(SOFCs) is proposed in this paper.A new estimator is proposed in a simple form to solve the modeling uncertainty,and express the control of SOFCs as an optimal control ***,a new system thermoelectric parameter tracking control strategy is *** strategy uses data-driven to build the recursive neural network model of the system thermoelectric *** order to ensure convergence,an adaptive law based on parameter estimation error is proposed to update the weights of neural networks *** simulation results show that the data-driven model can effectively identify the power generation process of solid oxide fuel cell *** the same time,the control method can realize the optimal tracking control of power,voltage,stack temperature and afterburner temperature with limited input.

关键词： Solid oxide fuel cell adaptive optimal control adaptive dynamic programming system thermoelectric performance

来源：评论

学校读者我要写书评

暂无评论

Self-Learning Optimal Control with Performance Analysis using Event-Triggered adaptive dynamic programming 21

Self-Learning Optimal Control with Performance Analysis usin...

引用

5th International Conference on Crowd Science and Engineering

作者： Ziyang Wang Joonhyup Lee Xiao Sun Yueting Chai Yi Liu National Engineering Laboratory for E-Commerce Technologies Tsinghua University China

ISBN: (纸本)9781450395540

In this paper, a self-learning event-triggered algorithm is proposed to solve optimal control problems for discrete-time systems. Not quite the same as the time-triggered algorithm, the developed controller is only updated when events occur. To obtain a desired performance, we develop novel triggering conditions with adjustable parameters. Therefore, a near-optimal performance can be obtained as required. Meanwhile, the system is proved to be asymptotically stable based on the developed algorithm. Besides, two neural networks are utilized to implement the algorithm. Furthermore, we test the developed algorithm with different parameter settings, and the impact of parameter settings on the performance is illustrated. Finally, we test both the time-triggered and event-triggered algorithms for comparison, which demonstrates that we obtain comparable performance with reduced computing cost.

关键词： neural network reinforcement learning event-triggered control adaptive dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Optimal Tracking Control For A Two-link Robotic Manipulator Via adaptive dynamic programming

Optimal Tracking Control For A Two-link Robotic Manipulator ...

引用

Chinese Automation Congress (CAC)

作者： Yang, Jiae Wang, Yujia Wang, Tong Yu, Xinghu Harbin Inst Technol Res Inst Intelligent Control & Syst Harbin Peoples R China Harbin Inst Technol Ningbo Inst Intelligent Equipment Technol Ningbo Peoples R China

ISBN: (纸本)9781728176871

An optimal tracking control strategy for a two-link robotic manipulator by using adaptive dynamic programming is investigated in this paper. Generally, system transformation is required when applying adaptive dynamic programming (ADP) technique to solve the issue of output tracking control design, which increases the difficulty and complexity in designing optimal output tracker. In order to overcome such obstacle, a novel designed cost function is employed such that the process of control design is simplified. Meanwhile, the optimal controller concerning the controlled system is obtained. The investigated novel optimal control strategy is conducted on a two-link manipulator control system, and simulation results prove that the presented scheme is efficient and feasible.

关键词： Robot manipulator Optimal tracking control adaptive dynamic programming Cost function

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：