检索结果-南通市图书馆

AUV Based Source Seeking with Estimated Gradients

在线全文

学校读者我要写书评

暂无评论

Journal of Systems Science & Complexity 2018年第1期31卷 262-275页

作者： LI Zhuo you keyou SONG Shiji Department of Automation and TNList Tsinghua University

This paper addresses the source seeking problems for an autonomous underwater vehicle（AUV） with the estimated gradients. The AUV is embedded with multiple sensors, which are only able to detect the signal strengths of the source with unknown distribution. To resolve this challenge,a sensor configuration is explicitly designed as a semicircle to estimate gradients of the signal ***, a controller is obtained via the estimated gradients to drive the AUV to approach the ***, an upper bound for the localization error is provided in terms of the radius of the semicircle and the signal distribution. Finally, the authors include a simulation example by applying the strategy to a Remote Environmental Monitoring Unit S（REMUS） for seeking the deepest point of a region of seabed in the South China Sea.

关键词： AUV dynamics gradient estimation source seeking

CONSENSUS DESIGN FOR CONTINUOUS-TIME MULTI-AGENT SYSTEMS WITH COMMUNICATION DELAY

在线全文

学校读者我要写书评

暂无评论

Journal of Systems Science & Complexity 2014年第4期27卷 701-711页

作者： WANG Zhenhua you keyou XU Juanjuan ZHANG Huanshui School of Control Science and Engineering Shandong University Department of Automation Tsinghua University

This paper investigates a consensus design problem for continuous-time first-order multiagent systems with uniform constant communication *** that the agent dynamic is unstable and the diagraph is undirected,sufficient conditions are derived to guarantee *** key technique is the adoption of historical input information in the ***,when agent's own historical input information is used in the protocol design,the consensus condition is constructed in terms of agent dynamic,communication delay,and the eigenratio of the network *** result is presented to validate the effectiveness of the theoretical result.

关键词： Communication delay consensus protocol eigenratio historical input information multi-agent systems.

Fully asynchronous distributed optimization with linear convergence over directed networks

在线全文

学校读者我要写书评

暂无评论

中山大学学报（自然科学版）（中英文） 2023年第5期62卷 1-23页

作者： SHA Xingyu ZHANG Jiaqi you keyou Department of Automation Beijing National Research Center for Information Science and TechnologyTsinghua UniversityBeijing 100084China Beijing Sankuai Online Technology Co. Ltd.Beijing 100102China

We study distributed optimization problems over a directed network,where nodes aim to minimize the sum of local objective functions via directed communications with *** algorithms are designed to solve it for synchronized or randomly activated implementation,which may create deadlocks in *** sharp contrast,we propose a fully asynchronous push-pull gradient(APPG) algorithm,where each node updates without waiting for any other node by using possibly delayed information from ***,we construct two novel augmented networks to analyze asynchrony and delays,and quantify its convergence rate from the worst-case point of ***,all nodes of APPG converge to the same optimal solution at a linear rate of O(λ^(k)) if local functions have Lipschitz-continuous gradients and their sum satisfies the Polyak-?ojasiewicz condition(convexity is not required),where λ ∈(0,1) is explicitly given and the virtual counter k increases by one when any node ***,the advantage of APPG over the synchronous counterpart and its linear speedup efficiency are numerically validated via a logistic regression problem.

关键词： fully asynchronous distributed optimization linear convergence Polyak-Łojasiewicz condition

Unsupervised learning of Dirichlet process mixture models with missing data

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2016年第1期59卷 161-174页

作者： Xunan ZHANG Shiji SONG Lei ZHU keyou you Cheng WU Department of Automation Tsinghua University China Ocean Mineral Resources R&D Association

This study presents a novel approach to unsupervised learning for clustering with missing *** first extend a finite mixture model to the infinite case by considering Dirichlet process mixtures, which can automatically determine the number of mixture components or clusters. Furthermore, we view the missing features as latent variables and compute the posterior distributions using the variational Bayesian expectation maximization algorithm, which optimizes the evidence lower bound on the complete-data log marginal likelihood. We demonstrate the performance on several artificial data sets with missing values. The experimental results indicate that the proposed method outperforms some classic imputation methods. We finally present an application to seabed hydrothermal sulfide color images analysis problem.

关键词： Dirichlet processes missing data clustering variational Bayesian image analysis

Opinion Dynamics in Relative-Opinion-Dependent Networks 36

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Opinion Dynamics in Relative-Opinion-Dependent Networks

第36届中国控制会议

作者： Zhuolin Wang Pei Xie keyou you Department of Automation and TNList Tsinghua University

ISBN: (纸本)9781538629185

This paper investigates the asymptotic behavior of opinion dynamics in relative-opinion-dependent networks,which is motivated by the observation that a social agent tends to accept opinions close to *** is,the interpersonal influence between social agents decreases gradually in their relative *** is interesting that this model can also provide a unified framework to study both the celebrated Degroot model and Deffuant-Weisbuch（DW） *** the social network is connected,we show that the Degroot model with opinion-dependent influence asymptotically reaches *** effect of the relativeopinion on the Friedkin-Johnsen（FJ） model is also illustrated via simulations.

关键词： Opinion dynamics relative-opinion-dependent networks Degroot model DW model FJ model

Distributed Vector Optimization Design for Multi-AUV Systems 36

在线全文

cnki会议

学校读者我要写书评

暂无评论

Distributed Vector Optimization Design for Multi-AUV Systems

第36届中国控制会议

作者： Zhuo Li keyou you Shiji Song Department of Automation Tsinghua Universityand TNList

ISBN: (纸本)9781538629185

This paper proposes a distributed controller to optimize a sum of state-dependent vector objective functions for multiAutonomous Underwater Vehicle（AUV） *** particular,each AUV has a local private objective and can only interact with its neighboring AUVs,which are described by undirected graphs,to measure relative *** 6 degree of freedom（DOF）AUV equations of motions,we design a distributed controller by integrating both the consensus and optimization algorithms to the dynamics of *** the undirected communication graph is connected,we adopt a Lyapunov functional candidate to show that each AUV eventually approaches the same optimal position with a desired ***,simulations with the Remote Environmental Measuring UnitS（REMUS） AUVs for source-seeking tasks are included to illustrate the effectiveness of the proposed controller.

关键词： Multi-AUV 6 DOF equation of motion distributed controller optimization

Policy evaluation for reinforcement learning over asynchronous multi-agent networks

在线全文

cnki会议

学校读者我要写书评

暂无评论

Policy evaluation for reinforcement learning over asynchrono...

第40届中国控制会议

作者： Xingyu Sha Jiaqi Zhang keyou you Department of Automation and BNRist Tsinghua University

This paper proposes a fully asynchronous algorithm for policy evaluation of multi-agent reinforcement learning over networks. Without any form of coordination, agents can communicate with neighbors and compute their local variables using（possibly） delayed information at any time. Thus, the proposed scheme fully takes advantage of the distributed setting. We prove that our method converges to a neighborhood of the optimum at a linear rate, showing the computational advantage by reducing the amount of synchronization. Numerical experiments show that our method is robust to straggler agents.

关键词： Multi-agent reinforcement learning multi-agent networks fully asynchronous updates policy evaluation

Stability of General Linear Dynamic Multi-Agent Systems under Switching Topologies with Positive Real Eigenvlues

在线全文

cnki会议

学校读者我要写书评

暂无评论

Engineering 2020年第6期6卷 688-694页

作者： Shengbo Eben Li Zhitao Wang Yang Zheng Diange Yang keyou you State Key Lab of Automotive Safety and Energy School of Vehicle and MobilityTsinghua UniversityBeijing 100084China Department of Engineering Science Balliol CollegeUniversity of OxfordOxford 0X13PJUK Department of Automation Tsinghua UniversityBeijing 100084China

The time-varying network topology can significantly affect the stability of multi-agent *** paper examines the stability of leader-follower multi-agent systems with general linear dynamics and switching network topologies,which have applications in the platooning of connected *** switching interaction topology is modeled as a class of directed graphs in order to describe the information exchange between multi-agent systems,where the eigenvalues of every associated matrix are required to be positive *** Hurwitz criterion and the Riccati inequality are used to design a distributed control law and estimate the convergence speed of the closed-loop system.A sufficient condition is provided for the stability of multi-agent systems under switching topologies.A common Lyapunov function is formulated to prove closed-loop stability for the directed network with switching *** result is applied to a typical cyber-physical system—that is,a connected vehicle platoon—which illustrates the effectiveness of the proposed method.

关键词： Stability Multi-agent system Switching topologies Common Lyapunov function

Velocity Regulation for Automatic Train Operation via Meta-Reinforcement Learning

在线全文

学校读者我要写书评

暂无评论

Velocity Regulation for Automatic Train Operation via Meta-R...

第三十九届中国控制会议

作者： Feiran Zhao keyou you Yunxin Fan Gang Yan Department of Automation and BNRist Tsinghua University CRRC Zhuzhou Locomotive Co.Ltd.

In this paper, we consider the velocity regulation for automatic train operation to maintain the velocity of the train at a desired value. Due to complicated railway conditions and uncertain dynamics of the system, the problem cannot be well solved by most of the model-based controllers. To this purpose, we formulate the velocity regulation problem as a sequence of stationary Markov decision processes（MDP） with unknown transition probabilities. Based on the meta-learning framework,we propose a model-free algorithm to learn an adaptive controller, which only requires a "small" amount of sampled data from the corresponding MDP. We illustrate with simulations that our model-free controller performs well and can well adapt to the dynamical environments.

关键词： Velocity regulation Markov decision procession Reinforcement learning Meta-learning Neural network