检索结果-内蒙古大学图书馆

ieee Conference on Decision and Control

作者： Tyler Summers Karthik Ganapathy Iman Shames Mathias Hudoba de Badyn Control Optimization and Networks Lab University of Texas at Dallas Australian National University Automatic Control Laboratory ETH Zürich

ISBN: (数字)9781665467612

ISBN: (纸本)9781665467629

We formulate a general mathematical framework for self-tuning network control architecture design. This problem involves jointly adapting the locations of active sensors and actuators in the network and the feedback control policy to all available information about the time-varying network state and dynamics to optimize a performance criterion. We propose a general solution structure analogous to the classical self-tuning regulator from adaptive control. We show that a special case with full-state feedback can be solved in principle with dynamic programming, and in the linear quadratic setting the optimal cost functions and policies are piecewise quadratic and piecewise linear, respectively. For large networks where exhaustive architecture search is prohibitive, we describe a greedy heuristic for joint architecture-policy design. We demonstrate in numerical experiments that self-tuning architectures can provide dramatically improved performance over fixed architectures. Our general formulation provides an extremely rich and challenging problem space with opportunities to apply a wide variety of approximation methods from stochastic control, system identification, reinforcement learning, and static architecture design.

关键词： Actuators reinforcement learning Aerospace electronics Cost function Control systems Sensors Feedback control

来源：评论

学校读者我要写书评

暂无评论

Application of reinforcement learning to The Orientation and Position Control of A 6 Degrees of Freedom Robotic Manipulator

Application of Reinforcement Learning to The Orientation and...

引用

ieee Latin American Robotics symposium, LARS

作者： Felipe Rigueira Campos Aline Xavier Fidêncio Jacó Domingues Gustavo Pessin Gustavo Freitas Programa de Pós-Graduação em Instrumentação Controle e Automação de Processos de Mineração Universidade Federal de Ouro Preto e Instituto Tecnológico Vale Ouro Preto MG Brazil Laboratório de Robótica Controle e Instrumentação Instituto Tecnológico Vale Ouro Preto MG Brazil Faculty of Electrical Engineering and Information Technology Ruhr-University Bochum Germany Departamento de Computação Universidade Federal de Ouro Preto Ouro Preto MG Brazil Departamento de Engenharia Elétrica Universidade Federal de Minas Gerais Belo Horizonte MG Brazil

ISBN: (数字)9781665462808

ISBN: (纸本)9781665462815

Applications with autonomous robots play an important role in the industry and in everyday life. Among them, the activities of manipulating and moving objects are highlighted by the wide variety of possible applications. These activities in static and known environments can be implemented through logic planned by the developer, but this is not feasible in dynamic environments. Machine learning (ML) techniques such as reinforcement learning (RL) algorithms have sought to replace the pre-defined programming by teaching the robot how to act. This paper presents the implementation of two RL algorithms, Deep Deterministic Policy Gradient (DDPG) and Proximal Policy Optimization (PPO), for orientation and position control of a 6-degree-of-freedom (6-DoF) robotic manipulator. The results demonstrated that the DDPG have a faster learning convergence in simpler activities, but if the complexity of the problem increases, it might not obtain a satisfactory behavior. On the other hand, PPO can solve more complex problems but it limits the convergence rate to the best result in order to avoid learning instability.

关键词： Service robots Position control reinforcement learning Power system stability Manipulators Stability analysis Behavioral sciences

来源：评论

学校读者我要写书评

暂无评论

Neural-Network-Based Robust Control Schemes for Nonlinear Multiplayer Systems With Uncertainties via adaptive dynamic programming

引用

ieee TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2019年第3期49卷 579-588页

作者： Jiang, He Zhang, Huaguang Luo, Yanhong Han, Ji Northeastern Univ Coll Informat Sci & Engn Shenyang 110819 Liaoning Peoples R China

This paper investigates the robust control issues of nonlinear multiplayer systems by utilizing adaptive dynamic programming (ADP) methods and fills a gap in the ADP field, where actuator uncertainties for multiplayer systems are still not addressed. Two types of actuator uncertainties including bounded nonlinear perturbation and unknown constant actuator fault are taken into consideration. First, a data-driven reinforcement learning (RL) approach is derived to learn the optimal solutions of multiplayer nonzero-sum games. Then, based on the obtained optimal control policies, two robust control schemes are developed to handle these two different types of uncertainties, respectively, and the associated stability analysis is also provided. To implement the proposed iterative RL approach, a single neural network (NN) architecture with least-square-based updating law is given, which reduces the computation burden compared with the traditional dual NN architecture. Finally, two numerical examples are shown to test the feasibility of our proposed schemes.

关键词： adaptive dynamic programming (ADP) approximate dynamic programming neural network (NN) reinforcement learning (RL)

来源：评论

学校读者我要写书评

暂无评论

Deep Conservative Policy Iteration 34

Deep Conservative Policy Iteration

引用

34th AAAI Conference on Artificial Intelligence / 32nd Innovative Applications of Artificial Intelligence Conference / 10th AAAI symposium on Educational Advances in Artificial Intelligence

作者： Vieillard, Nino Pietquin, Olivier Geist, Matthieu Google Res Brain Team Mountain View CA 94043 USA

ISBN: (纸本)9781577358350

Conservative Policy Iteration (CPI) is a founding algorithm of Approximate dynamic programming (ADP). Its core principle is to stabilize greediness through stochastic mixtures of consecutive policies. It comes with strong theoretical guarantees, and inspired approaches in deep reinforcement learning (RL). However, CPI itself has rarely been implemented, never with neural networks, and only experimented on toy problems. In this paper, we show how CPI can be practically combined with deep RL with discrete actions, in an off-policy manner. We also introduce adaptive mixture rates inspired by the theory. We experiment thoroughly the resulting algorithm on the simple Cartpole problem, and validate the proposed method on a representative subset of Atari games. Overall, this work suggests that revisiting classic ADP may lead to improved and more stable deep RL algorithms.

关键词： reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

AutoScale: Energy Efficiency Optimization for Stochastic Edge Inference Using reinforcement learning 53

AutoScale: Energy Efficiency Optimization for Stochastic Edg...

引用

53rd Annual ieee/ACM International symposium on Microarchitecture (MICRO)

作者： Kim, Young Geun Wu, Carole-Jean Arizona State Univ Tempe AZ 85287 USA Facebook AI Menlo Pk CA USA

ISBN: (纸本)9781728173832

Deep learning inference is increasingly run at the edge. As the programming and system stack support becomes mature, it enables acceleration opportunities in a mobile system, where the system performance envelope is scaled up with a plethora of programmable co-processors. Thus, intelligent services designed for mobile users can choose between running inference on the CPU or any of the co-processors in the mobile system, and exploiting connected systems such as the cloud or a nearby, locally connected mobile system. By doing so, these services can scale out the performance and increase the energy efficiency of edge mobile systems. This gives rise to a new challenge-deciding when inference should run where. Such execution scaling decision becomes more complicated with the stochastic nature of mobile-cloud execution environment, where signal strength variation in the wireless networks and resource interference can affect real-time inference performance and system energy efficiency. To enable energy efficient deep learning inference at the edge, this paper proposes AutoScale, an adaptive and lightweight execution scaling engine built on the custom-designed reinforcement learning algorithm. It continuously learns and selects the most energy efficient inference execution target by considering characteristics of neural networks and available systems in the collaborative cloud-edge execution environment while adapting to stochastic runtime variance. Real system implementation and evaluation, considering realistic execution scenarios, demonstrate an average of 9.8x and 1.6x energy efficiency improvement over the baseline mobile CPU and cloud offloading, respectively, while meeting the real-time performance and accuracy requirements.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

The Voltage Regulation of Boost Converters Using Dual Heuristic programming 10

The Voltage Regulation of Boost Converters Using Dual Heuris...

引用

10th Annual Computing and Communication Workshop and Conference (CCWC)

作者： Saadatmand, Sepehr Kavousi, Mohammadamir Azizi, Sima Missouri Univ S&T Dept Elect & Comp Engn Rolla MO 65409 USA Univ Calif Riverside Dept Elect & Comp Engn Riverside CA 92521 USA

ISBN: (纸本)9781728137834

In this paper, a dual heuristic programming controller is proposed to control a boost converter. Conventional controllers such as proportional-integral-derivative (PID) or proportional-integral (PI) are designed based on the linearized small-signal model near the operating point. Therefore, the performance of the controller during start-up, load change, or input voltage variation is not optimal since the system model changes by varying the operating point. The dual heuristic programming controller optimally controls the boost converter by following the approximate dynamic programming. The advantage of the DHP is that the neural network-based characteristic of the proposed controller enables boost converters to easily cope with large disturbances. A DHP with a well-trained critic and action networks can perform as an optimal controller for the boost converter. To compare the effectiveness of the traditional PI-based and the DHP boost converter, the simulation results are provided.

关键词： adaptive critic design Boost converter DC-DC converters Model predictive controller Dual heuristic programming reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Model-free Based reinforcement learning Control Strategy of Aircraft Attitude Systems

Model-free Based Reinforcement Learning Control Strategy of ...

引用

Chinese Automation Congress (CAC)

作者： Huang, Dingcui Hu, Jiangping Peng, Zhinan Chen, Bo Hao, Mingrui Ghosh, Bijoy Kumar Univ Elect Sci & Technol China Sch Automat Engn Chengdu Peoples R China Syst Control & Intelligent Agent Cooperat Lab Sci & Technol Complex Beijing Peoples R China Texas Tech Univ Dept Math & Stat Lubbock TX 79409 USA

ISBN: (纸本)9781728176871

Traditional aircraft control algorithms have a strong dependence on system models, and are difficult to cope with the increasingly complex battlefield environment for intelligent aircrafts. In this paper, a model-free reinforcement learning is proposed to solve an attitude stabilization problem of an aircraft based online intelligent control strateu. The attitude control problem is firstly formulated as an optimal control problem, and then an adaptive dynamic programming (ADP) technology is applied to compute the corresponding nonlinear Hamilton-Jacobi-Bellman (HJB) equation. Then, an actor-critic neural network structure is established to learn the optimal controller online not requiring the information of the aircraft dynamics. The proposed intelligent control strategy enables the aircraft to adjust its attitude according to the actual mission targets and environments under the proposed online control strategy, so that autonomous learning and intelligent operation can be realized. Finally, simulation examples are presented to validate the proposed model-free based control strategy.

关键词： Model-Free Control reinforcement learning Aircraft Attitude System Actor-Critic Neural Network

来源：评论

学校读者我要写书评

暂无评论

Distributed Frequency Controller for MT-HVDC Systems Via adaptive dynamic programming 32

Distributed Frequency Controller for MT-HVDC Systems Via Ada...

引用

32nd Chinese Control And Decision Conference (CCDC)

作者： Hu, Zhongjie Liu, Zhi-Wei Hu, Xiong Chi, Ming Huazhong Univ Sci & Technol Coll Artificial Intelligence & Automat Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Minist Educ Key Lab Image Proc & Intelligent Control Wuhan 430074 Peoples R China

ISBN: (纸本)9781728158556

The exact mathematical models of multi-terminal high-voltage direct current (MT-HVDC) systems are hard to be obtained in the practical application of HVDC technology. To overcome this challenge, a model-free distributed frequency controller for MT-HVDC systems is proposed based on adaptive dynamic programming (ADP) in this paper. Specifically, for each AC area, only the local and neighboring sampling data of frequency is required without the MT-HVDC system model. Besides, the proposed controller is distributed which effectively balances the communication burden among the AC areas. Moreover, the proposed controller makes the connected AC areas share their power reserves via HVDC grids to compensate load disturbances so that the necessary power reserves can be downsized. Some simulations carried out on a five-terminal HVDC system evaluate the performance of the proposed controller.

关键词： MT-HVDC system distributed frequency controller adaptive dynamic programming (ADP) neural network reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Time-Varying Deep reinforcement Model Predictive Control for DC Power Converter Systems

A Time-Varying Deep Reinforcement Model Predictive Control f...

引用

ieee International symposium on Power Electronics for Distributed Generation Systems (PEDG)

作者： Milad Andalibi Mojtaba Hajihosseini Sam Teymoori Maryam Kargar Meysam Gheisarnejad Shiraz University Shiraz Iran Aarhus University Aarhus Denmark UCSC Santa Cruz California United States Islamic Azad University Tehran Iran Islamic Azad University Najafabad Iran

Today power converters, especially DC/DC converters, is of great importance in power electronics applications such as DC micro-grids (MGs). However, they have some limitation such as inability to handle constant power load (CPL) which results in instability problems in MGs. Thus, a controller with specific characters including, robustness and fast response to system dynamic is vital to address the unsteadiness. In this paper, an adaptive model prediction controller (AMPC) based on Deep reinforcement learning (DRL) is developed to tackle the de-stabilization problem. In the proposed AMPC controller, the controlling signal coefficient in each variable operation point is regarded as the adjustable controller parameter and adaptively designed by the learning ability of the Deep Q- Network (DQN) strategy, leading to a robust controlling approach. We have shown that our suggested smart controller for DC/DC converters feeding CPLs is robust and fast in dynamic response.

关键词： Adaptation models System dynamics reinforcement learning Predictive models Power electronics Robustness Distributed power generation

来源：评论

学校读者我要写书评

暂无评论

On closed-loop stability of model predictive controllers with learning costs 18

On closed-loop stability of model predictive controllers wit...

引用

18th European Control Conference (ECC)

作者： Beckenbach, Lukas Osinenko, Pavel Streif, Stefan Tech Univ Chemnitz Automat Control & Syst Dynam Lab D-09107 Chemnitz Germany

ISBN: (纸本)9783907144022

Model predictive controllers are commonly associated with a fixed running and/or terminal cost function. Recently, some possibilities of cost function adaptation inspired by reinforcement learning were investigated. The current study analyzes closed-loop stability of such controllers in a general way. It is shown what constraints on learned running and terminal cost are required for this sake. A particular feature of the suggested control scheme is that, unlike in some common model predictive controllers, an assumed local Lyapunov function does not have to satisfy a decay function not less than the running cost. Relation of the considered control scheme to a baseline model predictive controller and adaptive dynamic programming is discussed. In a case study, it is shown how different cost function adaptation schemes lead to different performance with respect to the infinite-horizon cost.

关键词： Cost functions

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：