Interaction is one of the crucial features of multl-agent systems, in which there are two kinds of interaction： agent-to-agent and human-to-agent. In order to unify the two kinds of interaction while designing multi-...Interaction is one of the crucial features of multl-agent systems, in which there are two kinds of interaction： agent-to-agent and human-to-agent. In order to unify the two kinds of interaction while designing multi-agent systems, this paper introduces Q language-a scenario description language for designing interaction among agents and humans. Based on Q, we propose an integrating interaction framework system for multi-agent coordination, in which Q scenarios are used to uniformly describe both kinds of interactions. Being in accordance to the characteristics of Q language, the Q-based framework makes the interaction process open and easily understood by the users. Additionally, it makes specific applications of multi-agent systems easy to be established by application designers. By applying agent negotiation in agent-mediated e-commerce and agent cooperation in interoperable information query on the Semantic Web, we illustrate how the presented framework for multi-agent coordination is implemented in concrete applications. At the same time, these two different applications also demonstrate usability of the presented framework and verify validity of Q language.展开更多
For estimation group competition and multiagent coordination strategy, this paper introduces a notion based on multiagent group. According to the control domain, it analyzes the multiagent strategy during competition ...For estimation group competition and multiagent coordination strategy, this paper introduces a notion based on multiagent group. According to the control domain, it analyzes the multiagent strategy during competition in the macroscopic. It has been adopted in robot soccer and result enunciates that our method does not depend on competition result. It can objectively quantitatively estimate coordination strategy.展开更多
This paper considers the consensus problem of dynamical multiple agents that communicate via a directed moving neighbourhood random network. Each agent performs random walk on a weighted directed network. Agents inter...This paper considers the consensus problem of dynamical multiple agents that communicate via a directed moving neighbourhood random network. Each agent performs random walk on a weighted directed network. Agents interact with each other through random unidirectional information flow when they coincide in the underlying network at a given instant. For such a framework, we present sufficient conditions for almost sure asymptotic consensus. Numerical examples are taken to show the effectiveness of the obtained results.展开更多
Successful coordination in multi-agent systems requires agents to achieve consensus.Previous works propose methods through information sharing,such as explicit information sharing via communication protocols or exchan...Successful coordination in multi-agent systems requires agents to achieve consensus.Previous works propose methods through information sharing,such as explicit information sharing via communication protocols or exchanging information implicitly via behavior prediction.However,these methods may fail in the absence of communication channels or due to biased modeling.In this work,we propose to develop dual-channel consensus(DuCC)via contrastive representation learning for fully cooperative multi-agent systems,which does not need explicit communication and avoids biased modeling.DuCC comprises two types of consensus:temporally extended consensus within each agent(inner-agent consensus)and mutual consensus across agents(inter-agent consensus).To achieve DuCC,we design two objectives to learn representations of slow environmental features for inner-agent consensus and to realize cognitive consistency as inter-agent consensus.Our DuCC is highly general and can be flexibly combined with various MARL algorithms.The extensive experiments on StarCraft multi-agent challenge and Google research football demonstrate that our method efficiently reaches consensus and performs superiorly to state-of-the-art MARL algorithms.展开更多
Wind-photovoltaic(PV)-hydrogen-storage multi-agent energy systems are expected to play an important role in promoting renewable power utilization and decarbonization.In this study,a coordinated operation method was pr...Wind-photovoltaic(PV)-hydrogen-storage multi-agent energy systems are expected to play an important role in promoting renewable power utilization and decarbonization.In this study,a coordinated operation method was proposed for a wind-PVhydrogen-storage multi-agent energy system.First,a coordinated operation model was formulated for each agent considering peer-to-peer power trading.Second,a coordinated operation interactive framework for a multi-agent energy system was proposed based on the theory of the alternating direction method of multipliers.Third,a distributed interactive algorithm was proposed to protect the privacy of each agent and solve coordinated operation strategies.Finally,the effectiveness of the proposed coordinated operation method was tested on multi-agent energy systems with different structures,and the operational revenues of the wind power,PV,hydrogen,and energy storage agents of the proposed coordinated operation model were improved by approximately 59.19%,233.28%,16.75%,and 145.56%,respectively,compared with the independent operation model.展开更多
Cooperative multi-agent reinforcement learning(MARL)is a key technology for enabling cooperation in complex multi-agent systems.It has achieved remarkable progress in areas such as gaming,autonomous driving,and multi-...Cooperative multi-agent reinforcement learning(MARL)is a key technology for enabling cooperation in complex multi-agent systems.It has achieved remarkable progress in areas such as gaming,autonomous driving,and multi-robot control.Empowering cooperative MARL with multi-task decision-making capabilities is expected to further broaden its application scope.In multi-task scenarios,cooperative MARL algorithms need to address 3 types of multi-task problems:reward-related multi-task,arising from different reward functions;multi-domain multi-task,caused by differences in state and action spaces,state transition functions;and scalability-related multi-task,resulting from the dynamic variation in the number of agents.Most existing studies focus on scalability-related multitask problems.However,with the increasing integration between large language models(LLMs)and multi-agent systems,a growing number of LLM-based multi-agent systems have emerged,enabling more complex multi-task cooperation.This paper provides a comprehensive review of the latest advances in this field.By combining multi-task reinforcement learning with cooperative MARL,we categorize and analyze the 3 major types of multi-task problems under multi-agent settings,offering more fine-grained classifications and summarizing key insights for each.In addition,we summarize commonly used benchmarks and discuss future directions of research in this area,which hold promise for further enhancing the multi-task cooperation capabilities of multi-agent systems and expanding their practical applications in the real world.展开更多
This paper addresses the consensus problem of nonlinear multi-agent systems subject to external disturbances and uncertainties under denial-ofservice(DoS)attacks.Firstly,an observer-based state feedback control method...This paper addresses the consensus problem of nonlinear multi-agent systems subject to external disturbances and uncertainties under denial-ofservice(DoS)attacks.Firstly,an observer-based state feedback control method is employed to achieve secure control by estimating the system's state in real time.Secondly,by combining a memory-based adaptive eventtriggered mechanism with neural networks,the paper aims to approximate the nonlinear terms in the networked system and efficiently conserve system resources.Finally,based on a two-degree-of-freedom model of a vehicle affected by crosswinds,this paper constructs a multi-unmanned ground vehicle(Multi-UGV)system to validate the effectiveness of the proposed method.Simulation results show that the proposed control strategy can effectively handle external disturbances such as crosswinds in practical applications,ensuring the stability and reliable operation of the Multi-UGV system.展开更多
With the boom in maritime activities,the need for highly reliable maritime communication is becoming urgent,which is an important component of 5G/6G communication networks.However,the bandwidth reuse characteristic of...With the boom in maritime activities,the need for highly reliable maritime communication is becoming urgent,which is an important component of 5G/6G communication networks.However,the bandwidth reuse characteristic of 5G/6G networks will inevitably lead to severe interference,resulting in degradation in the communication performance of maritime users.In this paper,we propose a safe deep reinforcement learning based interference coordination scheme to jointly optimize the power control and bandwidth allocation in maritime communication systems,and exploit the quality-of-service requirements of users as the risk value references to evaluate the communication policies.In particular,this scheme designs a deep neural network to select the communication policies through the evaluation network and update the parameters using the target network,which improves the communication performance and speeds up the convergence rate.Moreover,the Nash equilibrium of the interference coordination game and the computational complexity of the proposed scheme are analyzed.Simulation and experimental results verify the performance gain of the proposed scheme compared with benchmarks.展开更多
This paper investigates the challenges associated with Unmanned Aerial Vehicle (UAV) collaborative search and target tracking in dynamic and unknown environments characterized by limited field of view. The primary obj...This paper investigates the challenges associated with Unmanned Aerial Vehicle (UAV) collaborative search and target tracking in dynamic and unknown environments characterized by limited field of view. The primary objective is to explore the unknown environments to locate and track targets effectively. To address this problem, we propose a novel Multi-Agent Reinforcement Learning (MARL) method based on Graph Neural Network (GNN). Firstly, a method is introduced for encoding continuous-space multi-UAV problem data into spatial graphs which establish essential relationships among agents, obstacles, and targets. Secondly, a Graph AttenTion network (GAT) model is presented, which focuses exclusively on adjacent nodes, learns attention weights adaptively and allows agents to better process information in dynamic environments. Reward functions are specifically designed to tackle exploration challenges in environments with sparse rewards. By introducing a framework that integrates centralized training and distributed execution, the advancement of models is facilitated. Simulation results show that the proposed method outperforms the existing MARL method in search rate and tracking performance with less collisions. The experiments show that the proposed method can be extended to applications with a larger number of agents, which provides a potential solution to the challenging problem of multi-UAV autonomous tracking in dynamic unknown environments.展开更多
Studying the coupling coordination development of new energy vehicles(NEVs)and the ecological environment in China is helpful in promoting the development of NEVs in the country and is of great significance in promoti...Studying the coupling coordination development of new energy vehicles(NEVs)and the ecological environment in China is helpful in promoting the development of NEVs in the country and is of great significance in promoting high-quality development of new energy in China.This paper constructs an evaluation index system for the development of NEVs and the ecological environment.It uses game theory combining weighting model,particle swarm optimized projection tracking evaluation model,coupling coordination degree model,and machine learning algorithms to calculate and analyze the level of coupling coordination development of NEVs and the ecological environment in China from 2010 to 2021,and identifies the driving factors.The research results show that:(i)From 2010 to 2021,the development index of NEVs in China has steadily increased from 0.085 to 0.634,while the ecological environment level index significantly rose from 0.170 to 0.884,reflecting the continuous development of China in both NEVs and the ecological environment.(ii)From 2010 to 2012,the two systems—new energy vehicle(NEV)development and the ecological environment—were in a period of imbalance and decline.From 2013 to 2016,they underwent a transition period,and from 2017 to 2021,they entered a period of coordinated development showing a trend of benign and continuous improvement.By 2021,they reached a good level of coordination.(iii)Indicators such as the number of patents granted for NEVs,water consumption per unit of GDP,and energy consumption per unit of GDP are the main driving factors affecting the coupling coordination development of NEVs and the ecological environment in China.展开更多
One Yb(Ⅲ)-based coordination polymer,{[Yb(H_(2)dhtp)1.5(H_(2)O)_(4)]·3H_(2)O}n(1)(H_(4)dhtp=2,5-dihydroxytere-phthalic acid),was fabricated and structurally characterized by single-crystal X-ray diffraction,IR,p...One Yb(Ⅲ)-based coordination polymer,{[Yb(H_(2)dhtp)1.5(H_(2)O)_(4)]·3H_(2)O}n(1)(H_(4)dhtp=2,5-dihydroxytere-phthalic acid),was fabricated and structurally characterized by single-crystal X-ray diffraction,IR,powder X-ray diffraction,X-ray diffraction,and elemental analysis.Complex 1 displays a 1D chain structure,and belongs to P1 group.The solid-state luminescent spectrum of 1 showed an emission band with the maximum at 508 nm(λex=408 nm).It exhibited the emission characteristic of the H_(4)dhtp ligand.The fluorescence of 1 in water displayed the stron-gest intensity.In detecting various metal ions,adding Zr^(4+)led to a blue shift in fluorescence,accompanied by an increase in intensity,whereas the presence of Fe^(3+)resulted in a decrease in luminescence.The changes observed in the IR spectrum indicate an interaction between Fe^(3+)/Zr^(4+)and complex 1,resulting in the variation of luminescence properties.展开更多
Under the condition of solvothermal synthesis,the viologen ligand 1,1′-bis(3-carboxyphenyl)-(4,4′-bipyri-dine)dichloride(H_(2)bcbpy·2Cl)and KI are coordinated with the metal cadmium ions.A case of thermochromic...Under the condition of solvothermal synthesis,the viologen ligand 1,1′-bis(3-carboxyphenyl)-(4,4′-bipyri-dine)dichloride(H_(2)bcbpy·2Cl)and KI are coordinated with the metal cadmium ions.A case of thermochromic coor-dination polymer[Cd(bcbpy)I_(2)]·2H_(2)O(1)was constructed.Complex 1 displays a 1D chain structure and exhibits thermochromic behavior.Under different temperature stimulation,the complex(ground)slowly changed from green to yellow-green,and with the increase of temperature,the color of complex 1 gradually deepened,and finally became orange-yellow.Therefore,complex 1 was prepared as a thermochromic film.In addition,we also performed electrochemical tests on complex 1,which showed that the complex is a semiconductor material.CCDC:2391802.展开更多
We report five coordination polymers(CPs)based on fluorescent ligands[1,6-di(1H-imidazol-1-yl)pyrene(dip),9,10-di(1H-imidazol-1-yl)anthracene(dia)]and anionic ligands[cyclohexane-1,4-dicarboxylic acid(H_(2)cda),campho...We report five coordination polymers(CPs)based on fluorescent ligands[1,6-di(1H-imidazol-1-yl)pyrene(dip),9,10-di(1H-imidazol-1-yl)anthracene(dia)]and anionic ligands[cyclohexane-1,4-dicarboxylic acid(H_(2)cda),camphoric acid(H_(2)cpa)].In[Cd(dip)(cda)]·4H_(2)O}_(n)(1),the Cd^(2+)ions,acting as tetrahedral nodes,are linked by dipand cda^(2-)ligands with four Cd^(2+)ions into five-fold interpenetrating network array of topology of dia.In{[Cd(dip)(cpa)]·4H_(2)O}_(n)(2),the Cd^(2+)ions,acting as a 4-connector,are linked by cpa^(2-)and dip ligands into a 3D framework ofcds topology.In{[Ni(dia)_(2)Cl_(2)]·DMF}_(n)(3),the Ni^(2+)ion is linked by four dia ligands into a layer structure,and 1Dchannels of a cross-section of 1.35 nm×0.96 nm are formed.In{[Cd(dia)_(2)(H_(2)O)_(2)](NO_(3))_(2)·2DMSO}n(4),the dia ligandsconnected Cd^(2+)ions into a 2D layer,and 1D channels are formed between adjacent layers with a cross-section of0.87 nm×0.43 nm.In[Zn(dip)Cl_(2)]_(n)(5),the Zn^(2+)ion is linked by dip ligands into an infinite 1D chain.The infrared,thermal gravimetric,and fluorescent emission data were collected and analyzed for these coordination polymers.CCDC:2356055,1;2440075,2;2356057,3;2356057,4;2356059,5.展开更多
Six coordination polymers based on 9,10-di(pyridine-4-yl)-anthracene(DPA)and 1,6-di(1H-imidazol-1-yl)pyrene(DIP)were obtained by solvothermal reactions.{[Zn(DPA)Cl_(2)]·DMF·2H_(2)O}n(1)and{[Zn_(1.5)(DPA)_(1....Six coordination polymers based on 9,10-di(pyridine-4-yl)-anthracene(DPA)and 1,6-di(1H-imidazol-1-yl)pyrene(DIP)were obtained by solvothermal reactions.{[Zn(DPA)Cl_(2)]·DMF·2H_(2)O}n(1)and{[Zn_(1.5)(DPA)_(1.5)Cl_(3)]·5H_(2)O}n(2)are framework isomers,which both contain zigzag chains formed by DPA,Zn^(2+),and Cl-.The zigzag chains in 1 are further assembled by C—H…Cl interactions into layers,and these layers exhibit two different orientations,displaying a rare 2D to 3D interpenetration mode.The zigzag chains in 2 are parallelly arranged.{[Zn_(3)(DPA)_(3)Br_(6)]·2DMF·_(1.5)H_(2)O}n(3)is isostructural to 2.3 was obtained using ZnBr_(2)instead of ZnCl_(2).[M(DPA)(formate)_(2)(H_(2)O)_(2)]n[M=Co(4),Cu(5)]are isostructural,contain chain structures formed by DPA,Cu^(2+)/Co^(2+),and for-mate ions,which were formed in situ in the solvothermal reaction.{[Zn(DIP)_(2)Cl]ClO_(4)}n(6)contains a layer structure formed by DIP and Zn^(2+).Free DPA and DIP ligands exhibited high fluorescence at room temperature,and coordina-tion polymers 3 and 6 displayed enhanced fluorescent emissions.展开更多
Two new Mn(Ⅱ)coordination polymers,namely{[Mn_(2)(HL)(phen)_(3)(H_(2)O)_(2)]·7.5H_(2)O}_n(1)and[Mn_(4)(HL)_(2)(1,4-bib)_(3)(H_(2)O)_(2)]_n(2),were synthesized under hydrothermal conditions by using Mn(Ⅱ)ions an...Two new Mn(Ⅱ)coordination polymers,namely{[Mn_(2)(HL)(phen)_(3)(H_(2)O)_(2)]·7.5H_(2)O}_n(1)and[Mn_(4)(HL)_(2)(1,4-bib)_(3)(H_(2)O)_(2)]_n(2),were synthesized under hydrothermal conditions by using Mn(Ⅱ)ions and 6-(3',4'-dicarboxylphenoxy)-1,2,4-benzenetricarboxylic acid(H_(5)L)in the presence of N-auxiliary ligands 1,10-phenanthroline(phen)and1,4-bis(1H-imidazol-1-yl)benzene(1,4-bib).The structures of coordination polymers 1 and 2 were characterized by infrared spectroscopy,single-crystal X-ray diffraction,thermogravimetric analysis,and powder X-ray diffraction.Single-crystal X-ray diffraction reveals that 1 has a 1D chain structure based on binuclear Mn(Ⅱ)units,while 2 features a(3,8)-connected 3D network structure based on tetranuclear Mn(Ⅱ)units.Magnetic studies show that 1 and 2exhibit antiferromagnetic interactions between manganese ions.2 shows stronger antiferromagnetic interactions due to the shorter Mn…Mn distances within the tetranuclear manganese units.CCDC:2357601,1;2357602,2.展开更多
A novel coordination polymer(CP){[Cd_(2)(L)(1,4-bimb)_(1.5)(DMF)_(2)]·DMF}n(1)(H_(4)L=5,5'-[1,1'-biphenyl-4,4'-diylbis(oxy)]diisophthalic acid,1,4-bimb=1,4-bis(imidazole-1-ylmethyl)-benzene)has been d...A novel coordination polymer(CP){[Cd_(2)(L)(1,4-bimb)_(1.5)(DMF)_(2)]·DMF}n(1)(H_(4)L=5,5'-[1,1'-biphenyl-4,4'-diylbis(oxy)]diisophthalic acid,1,4-bimb=1,4-bis(imidazole-1-ylmethyl)-benzene)has been designed and synthesized through solvothermal reaction.Structural analysis shows that Cd(Ⅱ)is connected by H4L and 1,4-bimb to form a 2D network,and 1,4-bimb further expands the 2D network into a 3D framework.CP 1 can be used as an excellent fluorescence sensor for Fe^(3+)and 4-nitrophenol(4-NP),with low detection limits and good anti-interference.The detection limits of Fe^(3+)and 4-NP were 0.034 and 0.031μmol·L^(-1),respectively.In addition,the fluorescence quenching mechanism was studied.1 was successfully applied to determine Fe^(3+)and 4-NP content in the Yanhe River water sample.CCDC:2351092.展开更多
High-entropy alloy(HEA)nanoparticles(NPs)have attracted great attention in electrocatalysis due to their tailorable complex compositions and unique properties.Herein,we introduce Fe,Co,Ni,Cr and Mn into the metal-poly...High-entropy alloy(HEA)nanoparticles(NPs)have attracted great attention in electrocatalysis due to their tailorable complex compositions and unique properties.Herein,we introduce Fe,Co,Ni,Cr and Mn into the metal-polyphenol coordination system to prepare HEA NPs enclosed in N-doped carbon(FeCoNiCrMn)with great potential for catalyzing oxygen reduction reaction(ORR)and oxygen evolution reaction(OER).The unique high-entropy structural characteristics in FeCoNiCrMn facilitate effective interplay between metal species,leading to improved ORR(E_(1/2)=0.89 V)and OER(η=330 mV,j=10 mA·cm^(−2))activity.Additionally,FeCoNiCrMn exhibits excellent open-circuit voltage(1.523 V),power density(110 mW·cm^(−2))and long-term durability,outperforming Pt/C+IrO_(2) electrodes as a cathode catalyst in Zn-air batteries(ZABs).Such polyphenol-assisted alloying method broadens and simplifies the development of HEA electrocatalysts for high-performance ZABs.展开更多
A low-cost 1D cobalt-based coordination polymer(CP)[Co(BGPD)(DMSO)_(2)(H_(2)O)_(2)](Co-BD;H2BGPD=N,N'-bis(glycinyl)pyromellitic diimide;DMSO=dimethyl sulfoxide)was synthesized by a simple method,and its crystal st...A low-cost 1D cobalt-based coordination polymer(CP)[Co(BGPD)(DMSO)_(2)(H_(2)O)_(2)](Co-BD;H2BGPD=N,N'-bis(glycinyl)pyromellitic diimide;DMSO=dimethyl sulfoxide)was synthesized by a simple method,and its crystal structure was characterized.In a three-electrode system,Co-BD,as the electrode material for supercapacitors,achieved a specific capacitance of 830 F·g^(-1)at 1 A·g^(-1),equivalent to a specific capacity of 116.4 mAh·g^(-1),and exhibited high-rate capability,reaching 212 F·g^(-1)at 20 A·g^(-1).Impressively,Co-BD||rGO(reduced graphene oxide),representing an asymmetrical supercapacitor,owns a higher energy density of 14.2 Wh·kg^(-1)at 0.80 kW·kg^(-1),and an excellent cycle performance(After 4000 cycles at 1 A·g^(-1),the capacitance retention was up to 94%).CCDC:2418872.展开更多
Reaction of the non-substituted/substituted unsymmetric pinene-derived complex[Pt(N^C^N')Cl]with the aryl isocyanide 2,6-dimethylphenyl isocyanide(CNXyl)afforded a mixture of two isomeric species:the ionic complex...Reaction of the non-substituted/substituted unsymmetric pinene-derived complex[Pt(N^C^N')Cl]with the aryl isocyanide 2,6-dimethylphenyl isocyanide(CNXyl)afforded a mixture of two isomeric species:the ionic complex[Pt(κ^(3)-N^C^N')(CNXyl)]Cl([A]Cl)and the molecular complex[Pt(κ^(2)-N^C^N')(CNXyl)Cl](B).Isomer B was almost the dominating product.The structures of the isomer B derivatives bearing-CF_(3)and-Cl substituents on the pyridine ring of the pinene moiety(5B and 7B,respectively)have been confirmed by single-crystal X-ray diffraction,revealing a slightly distorted square planar geometry with trans-N_(N^C^N'),CNR configuration(The terminal N atom of theκ^(2)-N^C^N'ligand is trans to the isocyanide ligand CNXyl.).Isomer B is thermodynamically more stable,as confirmed by theoretical calculations.CCDC:2416415,5B;2416414,7B.展开更多
文摘Interaction is one of the crucial features of multl-agent systems, in which there are two kinds of interaction： agent-to-agent and human-to-agent. In order to unify the two kinds of interaction while designing multi-agent systems, this paper introduces Q language-a scenario description language for designing interaction among agents and humans. Based on Q, we propose an integrating interaction framework system for multi-agent coordination, in which Q scenarios are used to uniformly describe both kinds of interactions. Being in accordance to the characteristics of Q language, the Q-based framework makes the interaction process open and easily understood by the users. Additionally, it makes specific applications of multi-agent systems easy to be established by application designers. By applying agent negotiation in agent-mediated e-commerce and agent cooperation in interoperable information query on the Semantic Web, we illustrate how the presented framework for multi-agent coordination is implemented in concrete applications. At the same time, these two different applications also demonstrate usability of the presented framework and verify validity of Q language.
文摘For estimation group competition and multiagent coordination strategy, this paper introduces a notion based on multiagent group. According to the control domain, it analyzes the multiagent strategy during competition in the macroscopic. It has been adopted in robot soccer and result enunciates that our method does not depend on competition result. It can objectively quantitatively estimate coordination strategy.
文摘This paper considers the consensus problem of dynamical multiple agents that communicate via a directed moving neighbourhood random network. Each agent performs random walk on a weighted directed network. Agents interact with each other through random unidirectional information flow when they coincide in the underlying network at a given instant. For such a framework, we present sufficient conditions for almost sure asymptotic consensus. Numerical examples are taken to show the effectiveness of the obtained results.
基金supported by the Strategic Priority Research Program of the Chinese Academy of Sciences,China(No.XDA27030300)the Program for National Nature Science Foundation of China(62073324).
文摘Successful coordination in multi-agent systems requires agents to achieve consensus.Previous works propose methods through information sharing,such as explicit information sharing via communication protocols or exchanging information implicitly via behavior prediction.However,these methods may fail in the absence of communication channels or due to biased modeling.In this work,we propose to develop dual-channel consensus(DuCC)via contrastive representation learning for fully cooperative multi-agent systems,which does not need explicit communication and avoids biased modeling.DuCC comprises two types of consensus:temporally extended consensus within each agent(inner-agent consensus)and mutual consensus across agents(inter-agent consensus).To achieve DuCC,we design two objectives to learn representations of slow environmental features for inner-agent consensus and to realize cognitive consistency as inter-agent consensus.Our DuCC is highly general and can be flexibly combined with various MARL algorithms.The extensive experiments on StarCraft multi-agent challenge and Google research football demonstrate that our method efficiently reaches consensus and performs superiorly to state-of-the-art MARL algorithms.
基金supported by the Key Research and Development Program of Jiangsu Provincial Department of Science and Technology(BE2020081).
文摘Wind-photovoltaic(PV)-hydrogen-storage multi-agent energy systems are expected to play an important role in promoting renewable power utilization and decarbonization.In this study,a coordinated operation method was proposed for a wind-PVhydrogen-storage multi-agent energy system.First,a coordinated operation model was formulated for each agent considering peer-to-peer power trading.Second,a coordinated operation interactive framework for a multi-agent energy system was proposed based on the theory of the alternating direction method of multipliers.Third,a distributed interactive algorithm was proposed to protect the privacy of each agent and solve coordinated operation strategies.Finally,the effectiveness of the proposed coordinated operation method was tested on multi-agent energy systems with different structures,and the operational revenues of the wind power,PV,hydrogen,and energy storage agents of the proposed coordinated operation model were improved by approximately 59.19%,233.28%,16.75%,and 145.56%,respectively,compared with the independent operation model.
基金The National Natural Science Foundation of China(62136008,62293541)The Beijing Natural Science Foundation(4232056)The Beijing Nova Program(20240484514).
文摘Cooperative multi-agent reinforcement learning(MARL)is a key technology for enabling cooperation in complex multi-agent systems.It has achieved remarkable progress in areas such as gaming,autonomous driving,and multi-robot control.Empowering cooperative MARL with multi-task decision-making capabilities is expected to further broaden its application scope.In multi-task scenarios,cooperative MARL algorithms need to address 3 types of multi-task problems:reward-related multi-task,arising from different reward functions;multi-domain multi-task,caused by differences in state and action spaces,state transition functions;and scalability-related multi-task,resulting from the dynamic variation in the number of agents.Most existing studies focus on scalability-related multitask problems.However,with the increasing integration between large language models(LLMs)and multi-agent systems,a growing number of LLM-based multi-agent systems have emerged,enabling more complex multi-task cooperation.This paper provides a comprehensive review of the latest advances in this field.By combining multi-task reinforcement learning with cooperative MARL,we categorize and analyze the 3 major types of multi-task problems under multi-agent settings,offering more fine-grained classifications and summarizing key insights for each.In addition,we summarize commonly used benchmarks and discuss future directions of research in this area,which hold promise for further enhancing the multi-task cooperation capabilities of multi-agent systems and expanding their practical applications in the real world.
基金The National Natural Science Foundation of China(W2431048)The Science and Technology Research Program of Chongqing Municipal Education Commission,China(KJZDK202300807)The Chongqing Natural Science Foundation,China(CSTB2024NSCQQCXMX0052).
文摘This paper addresses the consensus problem of nonlinear multi-agent systems subject to external disturbances and uncertainties under denial-ofservice(DoS)attacks.Firstly,an observer-based state feedback control method is employed to achieve secure control by estimating the system's state in real time.Secondly,by combining a memory-based adaptive eventtriggered mechanism with neural networks,the paper aims to approximate the nonlinear terms in the networked system and efficiently conserve system resources.Finally,based on a two-degree-of-freedom model of a vehicle affected by crosswinds,this paper constructs a multi-unmanned ground vehicle(Multi-UGV)system to validate the effectiveness of the proposed method.Simulation results show that the proposed control strategy can effectively handle external disturbances such as crosswinds in practical applications,ensuring the stability and reliable operation of the Multi-UGV system.
文摘With the boom in maritime activities,the need for highly reliable maritime communication is becoming urgent,which is an important component of 5G/6G communication networks.However,the bandwidth reuse characteristic of 5G/6G networks will inevitably lead to severe interference,resulting in degradation in the communication performance of maritime users.In this paper,we propose a safe deep reinforcement learning based interference coordination scheme to jointly optimize the power control and bandwidth allocation in maritime communication systems,and exploit the quality-of-service requirements of users as the risk value references to evaluate the communication policies.In particular,this scheme designs a deep neural network to select the communication policies through the evaluation network and update the parameters using the target network,which improves the communication performance and speeds up the convergence rate.Moreover,the Nash equilibrium of the interference coordination game and the computational complexity of the proposed scheme are analyzed.Simulation and experimental results verify the performance gain of the proposed scheme compared with benchmarks.
基金supported by the National Natural Science Foundation of China(Nos.12272104,U22B2013).
文摘This paper investigates the challenges associated with Unmanned Aerial Vehicle (UAV) collaborative search and target tracking in dynamic and unknown environments characterized by limited field of view. The primary objective is to explore the unknown environments to locate and track targets effectively. To address this problem, we propose a novel Multi-Agent Reinforcement Learning (MARL) method based on Graph Neural Network (GNN). Firstly, a method is introduced for encoding continuous-space multi-UAV problem data into spatial graphs which establish essential relationships among agents, obstacles, and targets. Secondly, a Graph AttenTion network (GAT) model is presented, which focuses exclusively on adjacent nodes, learns attention weights adaptively and allows agents to better process information in dynamic environments. Reward functions are specifically designed to tackle exploration challenges in environments with sparse rewards. By introducing a framework that integrates centralized training and distributed execution, the advancement of models is facilitated. Simulation results show that the proposed method outperforms the existing MARL method in search rate and tracking performance with less collisions. The experiments show that the proposed method can be extended to applications with a larger number of agents, which provides a potential solution to the challenging problem of multi-UAV autonomous tracking in dynamic unknown environments.
基金Supported by the Postgraduate Research&Practice Innovation Program of Jiangsu Province(KYCX24_0102)the China Scholarship Council Program(202406190114)。
文摘Studying the coupling coordination development of new energy vehicles(NEVs)and the ecological environment in China is helpful in promoting the development of NEVs in the country and is of great significance in promoting high-quality development of new energy in China.This paper constructs an evaluation index system for the development of NEVs and the ecological environment.It uses game theory combining weighting model,particle swarm optimized projection tracking evaluation model,coupling coordination degree model,and machine learning algorithms to calculate and analyze the level of coupling coordination development of NEVs and the ecological environment in China from 2010 to 2021,and identifies the driving factors.The research results show that:(i)From 2010 to 2021,the development index of NEVs in China has steadily increased from 0.085 to 0.634,while the ecological environment level index significantly rose from 0.170 to 0.884,reflecting the continuous development of China in both NEVs and the ecological environment.(ii)From 2010 to 2012,the two systems—new energy vehicle(NEV)development and the ecological environment—were in a period of imbalance and decline.From 2013 to 2016,they underwent a transition period,and from 2017 to 2021,they entered a period of coordinated development showing a trend of benign and continuous improvement.By 2021,they reached a good level of coordination.(iii)Indicators such as the number of patents granted for NEVs,water consumption per unit of GDP,and energy consumption per unit of GDP are the main driving factors affecting the coupling coordination development of NEVs and the ecological environment in China.
文摘One Yb(Ⅲ)-based coordination polymer,{[Yb(H_(2)dhtp)1.5(H_(2)O)_(4)]·3H_(2)O}n(1)(H_(4)dhtp=2,5-dihydroxytere-phthalic acid),was fabricated and structurally characterized by single-crystal X-ray diffraction,IR,powder X-ray diffraction,X-ray diffraction,and elemental analysis.Complex 1 displays a 1D chain structure,and belongs to P1 group.The solid-state luminescent spectrum of 1 showed an emission band with the maximum at 508 nm(λex=408 nm).It exhibited the emission characteristic of the H_(4)dhtp ligand.The fluorescence of 1 in water displayed the stron-gest intensity.In detecting various metal ions,adding Zr^(4+)led to a blue shift in fluorescence,accompanied by an increase in intensity,whereas the presence of Fe^(3+)resulted in a decrease in luminescence.The changes observed in the IR spectrum indicate an interaction between Fe^(3+)/Zr^(4+)and complex 1,resulting in the variation of luminescence properties.
文摘Under the condition of solvothermal synthesis,the viologen ligand 1,1′-bis(3-carboxyphenyl)-(4,4′-bipyri-dine)dichloride(H_(2)bcbpy·2Cl)and KI are coordinated with the metal cadmium ions.A case of thermochromic coor-dination polymer[Cd(bcbpy)I_(2)]·2H_(2)O(1)was constructed.Complex 1 displays a 1D chain structure and exhibits thermochromic behavior.Under different temperature stimulation,the complex(ground)slowly changed from green to yellow-green,and with the increase of temperature,the color of complex 1 gradually deepened,and finally became orange-yellow.Therefore,complex 1 was prepared as a thermochromic film.In addition,we also performed electrochemical tests on complex 1,which showed that the complex is a semiconductor material.CCDC:2391802.
文摘We report five coordination polymers(CPs)based on fluorescent ligands[1,6-di(1H-imidazol-1-yl)pyrene(dip),9,10-di(1H-imidazol-1-yl)anthracene(dia)]and anionic ligands[cyclohexane-1,4-dicarboxylic acid(H_(2)cda),camphoric acid(H_(2)cpa)].In[Cd(dip)(cda)]·4H_(2)O}_(n)(1),the Cd^(2+)ions,acting as tetrahedral nodes,are linked by dipand cda^(2-)ligands with four Cd^(2+)ions into five-fold interpenetrating network array of topology of dia.In{[Cd(dip)(cpa)]·4H_(2)O}_(n)(2),the Cd^(2+)ions,acting as a 4-connector,are linked by cpa^(2-)and dip ligands into a 3D framework ofcds topology.In{[Ni(dia)_(2)Cl_(2)]·DMF}_(n)(3),the Ni^(2+)ion is linked by four dia ligands into a layer structure,and 1Dchannels of a cross-section of 1.35 nm×0.96 nm are formed.In{[Cd(dia)_(2)(H_(2)O)_(2)](NO_(3))_(2)·2DMSO}n(4),the dia ligandsconnected Cd^(2+)ions into a 2D layer,and 1D channels are formed between adjacent layers with a cross-section of0.87 nm×0.43 nm.In[Zn(dip)Cl_(2)]_(n)(5),the Zn^(2+)ion is linked by dip ligands into an infinite 1D chain.The infrared,thermal gravimetric,and fluorescent emission data were collected and analyzed for these coordination polymers.CCDC:2356055,1;2440075,2;2356057,3;2356057,4;2356059,5.
文摘Six coordination polymers based on 9,10-di(pyridine-4-yl)-anthracene(DPA)and 1,6-di(1H-imidazol-1-yl)pyrene(DIP)were obtained by solvothermal reactions.{[Zn(DPA)Cl_(2)]·DMF·2H_(2)O}n(1)and{[Zn_(1.5)(DPA)_(1.5)Cl_(3)]·5H_(2)O}n(2)are framework isomers,which both contain zigzag chains formed by DPA,Zn^(2+),and Cl-.The zigzag chains in 1 are further assembled by C—H…Cl interactions into layers,and these layers exhibit two different orientations,displaying a rare 2D to 3D interpenetration mode.The zigzag chains in 2 are parallelly arranged.{[Zn_(3)(DPA)_(3)Br_(6)]·2DMF·_(1.5)H_(2)O}n(3)is isostructural to 2.3 was obtained using ZnBr_(2)instead of ZnCl_(2).[M(DPA)(formate)_(2)(H_(2)O)_(2)]n[M=Co(4),Cu(5)]are isostructural,contain chain structures formed by DPA,Cu^(2+)/Co^(2+),and for-mate ions,which were formed in situ in the solvothermal reaction.{[Zn(DIP)_(2)Cl]ClO_(4)}n(6)contains a layer structure formed by DIP and Zn^(2+).Free DPA and DIP ligands exhibited high fluorescence at room temperature,and coordina-tion polymers 3 and 6 displayed enhanced fluorescent emissions.
文摘Two new Mn(Ⅱ)coordination polymers,namely{[Mn_(2)(HL)(phen)_(3)(H_(2)O)_(2)]·7.5H_(2)O}_n(1)and[Mn_(4)(HL)_(2)(1,4-bib)_(3)(H_(2)O)_(2)]_n(2),were synthesized under hydrothermal conditions by using Mn(Ⅱ)ions and 6-(3',4'-dicarboxylphenoxy)-1,2,4-benzenetricarboxylic acid(H_(5)L)in the presence of N-auxiliary ligands 1,10-phenanthroline(phen)and1,4-bis(1H-imidazol-1-yl)benzene(1,4-bib).The structures of coordination polymers 1 and 2 were characterized by infrared spectroscopy,single-crystal X-ray diffraction,thermogravimetric analysis,and powder X-ray diffraction.Single-crystal X-ray diffraction reveals that 1 has a 1D chain structure based on binuclear Mn(Ⅱ)units,while 2 features a(3,8)-connected 3D network structure based on tetranuclear Mn(Ⅱ)units.Magnetic studies show that 1 and 2exhibit antiferromagnetic interactions between manganese ions.2 shows stronger antiferromagnetic interactions due to the shorter Mn…Mn distances within the tetranuclear manganese units.CCDC:2357601,1;2357602,2.
文摘A novel coordination polymer(CP){[Cd_(2)(L)(1,4-bimb)_(1.5)(DMF)_(2)]·DMF}n(1)(H_(4)L=5,5'-[1,1'-biphenyl-4,4'-diylbis(oxy)]diisophthalic acid,1,4-bimb=1,4-bis(imidazole-1-ylmethyl)-benzene)has been designed and synthesized through solvothermal reaction.Structural analysis shows that Cd(Ⅱ)is connected by H4L and 1,4-bimb to form a 2D network,and 1,4-bimb further expands the 2D network into a 3D framework.CP 1 can be used as an excellent fluorescence sensor for Fe^(3+)and 4-nitrophenol(4-NP),with low detection limits and good anti-interference.The detection limits of Fe^(3+)and 4-NP were 0.034 and 0.031μmol·L^(-1),respectively.In addition,the fluorescence quenching mechanism was studied.1 was successfully applied to determine Fe^(3+)and 4-NP content in the Yanhe River water sample.CCDC:2351092.
基金supported by the Fundamental Research Funds for the Central Universities(No.22120230104).
文摘High-entropy alloy(HEA)nanoparticles(NPs)have attracted great attention in electrocatalysis due to their tailorable complex compositions and unique properties.Herein,we introduce Fe,Co,Ni,Cr and Mn into the metal-polyphenol coordination system to prepare HEA NPs enclosed in N-doped carbon(FeCoNiCrMn)with great potential for catalyzing oxygen reduction reaction(ORR)and oxygen evolution reaction(OER).The unique high-entropy structural characteristics in FeCoNiCrMn facilitate effective interplay between metal species,leading to improved ORR(E_(1/2)=0.89 V)and OER(η=330 mV,j=10 mA·cm^(−2))activity.Additionally,FeCoNiCrMn exhibits excellent open-circuit voltage(1.523 V),power density(110 mW·cm^(−2))and long-term durability,outperforming Pt/C+IrO_(2) electrodes as a cathode catalyst in Zn-air batteries(ZABs).Such polyphenol-assisted alloying method broadens and simplifies the development of HEA electrocatalysts for high-performance ZABs.
文摘A low-cost 1D cobalt-based coordination polymer(CP)[Co(BGPD)(DMSO)_(2)(H_(2)O)_(2)](Co-BD;H2BGPD=N,N'-bis(glycinyl)pyromellitic diimide;DMSO=dimethyl sulfoxide)was synthesized by a simple method,and its crystal structure was characterized.In a three-electrode system,Co-BD,as the electrode material for supercapacitors,achieved a specific capacitance of 830 F·g^(-1)at 1 A·g^(-1),equivalent to a specific capacity of 116.4 mAh·g^(-1),and exhibited high-rate capability,reaching 212 F·g^(-1)at 20 A·g^(-1).Impressively,Co-BD||rGO(reduced graphene oxide),representing an asymmetrical supercapacitor,owns a higher energy density of 14.2 Wh·kg^(-1)at 0.80 kW·kg^(-1),and an excellent cycle performance(After 4000 cycles at 1 A·g^(-1),the capacitance retention was up to 94%).CCDC:2418872.
文摘Reaction of the non-substituted/substituted unsymmetric pinene-derived complex[Pt(N^C^N')Cl]with the aryl isocyanide 2,6-dimethylphenyl isocyanide(CNXyl)afforded a mixture of two isomeric species:the ionic complex[Pt(κ^(3)-N^C^N')(CNXyl)]Cl([A]Cl)and the molecular complex[Pt(κ^(2)-N^C^N')(CNXyl)Cl](B).Isomer B was almost the dominating product.The structures of the isomer B derivatives bearing-CF_(3)and-Cl substituents on the pyridine ring of the pinene moiety(5B and 7B,respectively)have been confirmed by single-crystal X-ray diffraction,revealing a slightly distorted square planar geometry with trans-N_(N^C^N'),CNR configuration(The terminal N atom of theκ^(2)-N^C^N'ligand is trans to the isocyanide ligand CNXyl.).Isomer B is thermodynamically more stable,as confirmed by theoretical calculations.CCDC:2416415,5B;2416414,7B.
