检索结果-南通市图书馆

Distributed multi-agent temporal-difference learning with full neighbor information

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Control Theory and Technology 2020年第4期18卷 379-389页

作者： zhinan peng Jiangping Hu Rui Luo Bijoy K.Ghosh School of Automation Engineering University of Electronic Science and Technology of ChinaChengdu 611731SichuanChina Department of Mathematics and Statistics Texas Tech UniversityLubbockTX 79409-1042USA

This paper presents a novel distributed multi-agent temporal-difference learning framework for value function approximation,which alows agents using all the neighbor information instead of the information from only one *** full neighbor information,the proposed framework(1)has a faster convergence rate,and(2)is more robust compared to the state of-the art *** we propose a distributed multi-agent discounted temporal dfferene algorithm and a distributed muli-agent average cost temporal diference leaming algorithm based on th ***,the two proposed algorthms'theoretical convergence proofs are *** simulation resuts show that our proposed algorihms are superior to the gossip-based algorithm in convergence speed,robustness to noise and time-varying network topology.

关键词： Distributed algorithm Reinforcement learning Temprel-lifferene learning Multi-agent systems

Optimal containment control of continuous-time multi-agent systems with unknown disturbances using data-driven approach

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2020年第10期63卷 270-272页

作者： zhinan peng Jiefu ZHANG Jiangping HU Rui HUANG Bijoy Kumar GHOSH School of Automation Engineering University of Electronic Science and Technology of China Department of Mathematics and Statistics Texas Tech University

Dear editor,Containment control problems of multi-agent systems (MASs) have attracted considerable attention in recent years owing to their widespread use in applications pertaining to spacecraft control,sensor networ... 详细信息

关键词： High-order multi-agent systems unmatched disturbances Containment control critic- actor network model transformation

Non-fragile memory sampled-data LFC for semi-Markov jump interconnected multi-area power systems: A looped functional approach

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Non-fragile memory sampled-data LFC for semi-Markov jump int...

第43届中国控制会议

作者： Yulong Li Xiaoqing Li zhinan peng Kaibo Shi School of Electrical Engineering and Automation Hefei University of Technology School of Automation Engineering University of Electronic Science and Technology of China School of Information Science and Engineering Chengdu University

The problem of memory sampled-data load frequency control(LFC) for semi-Markov jump interconnected multi-area power systems(IMAPSs) has been elaborately presented in this work. Firstly, the semi-Markov jump process(MJP) is introduced to model the dynamical random mutations of the IMAPSs. Subsequently, a memory sampled-data LFC method concerning signal transmission delay caused by geographical isolation for the semi-Markov jump IMPASs is accordingly delivered. By considering the parameter fluctuations in practical operation, the non-fragile issue for the proportional-integral(PI) controller is included to meet the actual operation scenario. After that, a mode-dependent two-side looped Lyapunov–Krasovskii functional is properly adopted to acquire enhanced flexibility of the derived sufficient condition in the shape of linear matrix inequalities(LMIs), which guarantees the stochastic stability with prescribed H∞ performance. Finally, a numerical example is exhibited to validate the effectiveness of the developed control strategy.

关键词： Semi-Markov jump interconnected multi-area power systems(IMAPSs) memory sampled-data load frequency control(LFC) non-fragile proportional-integral(PI) strategy

Optimal Tracking Control of Heterogeneous Multi-agent Systems with Switching Topology via Actor-Critic Neural Networks

在线全文

cnki会议

学校读者我要写书评

暂无评论

Optimal Tracking Control of Heterogeneous Multi-agent System...

第37届中国控制会议

作者： zhinan peng Jiangping Hu Bijoy K.Ghosh School of Automation Engineering University of Electronic Science and Technology of China Department of Mathematics and Statistics Texas Tech University

In this paper, an optimal tracking control problem is solved for high-order heterogeneous multi-agent systems with time-varying interaction networks within the framework of reinforcement learning. First, the optimal tracking control problem is formulated as a leader-follower multi-agent system. Second, a policy iteration based adaptive dynamic programming（ADP） algorithm is proposed to compute the performance index and the control law. Furthermore, the convergence to the optimal solutions is analyzed for the proposed algorithm. Third, an actor-critic neural network is applied to approximate the iterative performance index function and the control law, which implement the policy iteration algorithm online without using the knowledge of the system dynamics. Finally, some simulation results are presented to demonstrate the proposed optimal tracking control strategy.

关键词： Optimal tracking control multi-agent systems adaptive dynamic programming actor-critic neural network

Relative Angle Measurements Based Shape-Similar Formation for Mobile Robots with Input Time-Delay

在线全文

cnki会议

学校读者我要写书评

暂无评论

Relative Angle Measurements Based Shape-Similar Formation fo...

第三十九届中国控制会议

作者： Rui Luo zhinan peng Jiangping Hu Bijoy Kumar Ghosh School of Automation Engineering University of Electronic Science and Technology of China Department of Mathematics and Statistics Texas Tech University

This paper focuses on the infinitesimally shape-similar formation problem for a group of robots subject to input time delay. We assume that each robot is only equipped with bearing sensors. For the implementation of infinitesimally shapesimilar motion, the shape-similarity matrix（SSM） is introduced by using bearing angle measurement（BAM）. It is shown that the SSM only can maintain the angle-consistent but not keeping shape of formation along trajectories. Further, some conditions of selecting angles are given to guarantee the shape-preservation of formation on the basis of angle-consistent. To apply the above results, an optimization formation task is established for a group of robots with single integrator dynamics subject to input time-delay. This task aims at maintaining shape of formation as well as achieving position tracking. Then a shape-similar-based optimal controller is obtained by using Lagrangian multiplier method. Finally, a simulation scenario on a quadrilateral formation of four robots is carried out to demonstrate the proposed shape-similar based approaches.

关键词： Infinitesimally shape-similar formation Bearing angle measurement Shape-similar matrix Input time delay

Deterministic Learning-based Impedance Control for Human-Robot Interaction with Nonlinear Uncertain Dynamics

在线全文

cnki会议

学校读者我要写书评

暂无评论

Deterministic Learning-based Impedance Control for Human-Rob...

第43届中国控制会议

作者： Jingting Zhang zhinan peng Chaobin Zou Rui Huang Hong Cheng School of Automation Engineering University of Electronic Science And Technology of China

In this paper, we propose a Deterministic Learning-based impedance control scheme for human-robot interaction(HRI), which can deal with the uncertain nonlinear dynamics resulted by the unpredictable behavior in HRI process. A dynamicslearning-based impedance controller is proposed by embedding an adaptive radial basis function neural network(RBF NN) model to online learn the uncertain nonlinear dynamics of the robot from the closed-loop control during HRI process. With the proposed approach, the uncertain nonlinear dynamics of the robot can be accurated learned by guaranteeing the NN weights to converge to their optimal values, and the associated knowledge can be obtained and stored by a constant RBF NN model. A new experiencebased controller is then developed with the constant NN model, which can recall the stored knowledge whenever the same/similar control situation recurs, thereby leadning to high control performance with less computation efforts.

关键词： Human-Robot Interaction Impedance Control Dynamics Learning Nonlinear Uncertain System Deterministic Learning

Optimal Walking Assistance Control of Lower Limb Exoskeleton Using Adaptive Learning Approach

在线全文

cnki会议

学校读者我要写书评

暂无评论

Optimal Walking Assistance Control of Lower Limb Exoskeleton...

第三十九届中国控制会议

作者： zhinan peng Rui Luo Jiangping Hu Bijoy Kumar Ghosh Sing Kiong Nguang School of Automation Engineering University of Electronic Science and Technology of China Department of Mathematics and Statistics Texas Tech University Department of Electrical Computer and Software Engineering University of Auckland

In this paper, an adaptive reinforcement learning（RL） based controller is developed to solve assistance control problem for Lower Limb Exoskeleton（LLE） to aid hemiplegic individuals in walking. The communication interaction relation between both two lower-limbs and patient’s unaffected leg is modelled in the context of leader-follower（LF） framework. The walking assistance control problem of LLE with patients is converted to optimal control problem. To handle the optimal control problem,a discounted cost function is designed in terms of the local tracking error, and then a policy iteration algorithm（PI） is proposed to generate an optimal control policy, followed by the convergence analysis of the presented algorithm. Further, in order to improve the adaption of the controller to different patients, on the basis of the PI algorithm, an actor-critic-based neural network（AC/NN）architecture is introduced to implement the presented control method in an online-learning fashion. Finally, simulation scenarios are established to test the effectiveness of the proposed walking assistance control approaches.

关键词： Optimal walking assistance control Lower limb exoskeleton Reinforcement learning Actor-critic network

下肢助行外骨骼机器人起坐动作下的生物力学分析

在线全文

cnki会议

学校读者我要写书评

暂无评论

机器人 2021年第4期43卷 414-423页

作者：施柯丞黄瑞程洪邱静彭知南尹文泽电子科技大学自动化工程学院四川成都610054 电子科技大学机械与电气工程学院四川成都610054

探讨了下肢外骨骼起坐动作辅助过程中截瘫患者对拐杖产生大压力的原因和解决方法,通过建立人-外骨骼耦合动力学模型,将患者对拐杖的压力作为一个外力引入模型中,探究了踝关节刚度与拐杖压力的关系.为了验证踝关节刚度对减小拐杖压力的影... 详细信息

探讨了下肢外骨骼起坐动作辅助过程中截瘫患者对拐杖产生大压力的原因和解决方法,通过建立人-外骨骼耦合动力学模型,将患者对拐杖的压力作为一个外力引入模型中,探究了踝关节刚度与拐杖压力的关系.为了验证踝关节刚度对减小拐杖压力的影响,设计了外骨骼起坐验证实验,通过调整AIDER下肢助行外骨骼系统踝关节的刚度,观察受试者在整个起坐动作过程中的平均能量损耗和鞋底压力中心位置变化.实验结果显示,调整踝关节刚度后受试者的平均能量损耗降低了45.18%,稳定性提高了34.99%,这表明适当的踝关节刚度可以在下肢助行外骨骼辅助的起坐动作过程中有效降低患者对拐杖的压力.

关键词：起坐动作人机耦合动力学模型踝关节刚度下肢助行外骨骼截瘫患者

维普期刊数据库

Gut microbiota have blood types as human

在线全文

学校读者我要写书评

暂无评论

Science Bulletin 2018年第20期63卷 1311-1313页

作者： Qizheng Wu Hui Zhong Yafei Zhai Yanjiong Jia zhinan Yin Min Chen Hengwen Yang peng George Wang The State Key Laboratory of Microbial Technology and National Glycoengineering Research Center Shandong University The First Affiliated Hospital Biomedical Translational Research Institute Guangdong Province Key Laboratory of Molecular Immunology and Antibody EngineeringJinan University

ABO blood group system was firstly recognized by Landsteiner in 1900[1].Since then,the antigens of the ABO system (A,B and H determinants,respectively)have been shown to encompass complex carbohydrate structures [2].A and B antigen were synthesized by the sequential action of glycosyltransferases,with A and B glycosyltransferases catalyzing the addition of N-acetylgalactosamine and D-galactose to precursor H antigen,respectively. Group O individuals lack such transferase enzymes and consequently continue to express the basic H structure only [3,4].It is determined that approximate 2million ABH glycan antigen sites are presented on each red blood cell [5].Additionally,the ABH antigens are widely expressed in other human ceils and tissues,including the sensory neurons,epithelium,the vascular endothelium and platelets [6].

关键词： Gut microbiota blood types

维普期刊数据库