Aiming at the problems of low solution accuracy and high decision pressure when facing large-scale dynamic task allocation(DTA)and high-dimensional decision space with single agent,this paper combines the deep reinfor...Aiming at the problems of low solution accuracy and high decision pressure when facing large-scale dynamic task allocation(DTA)and high-dimensional decision space with single agent,this paper combines the deep reinforce-ment learning(DRL)theory and an improved Multi-Agent Deep Deterministic Policy Gradient(MADDPG-D2)algorithm with a dual experience replay pool and a dual noise based on multi-agent architecture is proposed to improve the efficiency of DTA.The algorithm is based on the traditional Multi-Agent Deep Deterministic Policy Gradient(MADDPG)algorithm,and considers the introduction of a double noise mechanism to increase the action exploration space in the early stage of the algorithm,and the introduction of a double experience pool to improve the data utilization rate;at the same time,in order to accelerate the training speed and efficiency of the agents,and to solve the cold-start problem of the training,the a priori knowledge technology is applied to the training of the algorithm.Finally,the MADDPG-D2 algorithm is compared and analyzed based on the digital battlefield of ground and air confrontation.The experimental results show that the agents trained by the MADDPG-D2 algorithm have higher win rates and average rewards,can utilize the resources more reasonably,and better solve the problem of the traditional single agent algorithms facing the difficulty of solving the problem in the high-dimensional decision space.The MADDPG-D2 algorithm based on multi-agent architecture proposed in this paper has certain superiority and rationality in DTA.展开更多
Complex and distributed systems are more and more associated with the application of WSN (Wireless Sensor Network) technology. The design of such applications presents important challenges and requires the assistance ...Complex and distributed systems are more and more associated with the application of WSN (Wireless Sensor Network) technology. The design of such applications presents important challenges and requires the assistance of several methodologies and tools. Multi-Agent systems (MAS) have been identified as one of the most suitable technologies to contribute to this domain due to their appropriateness for modeling distributed and autonomous complex systems. This work aims to contribute in the help of the design of WSN applications. The proposed architecture exploits the advantages of MAS for modeling WSN services, network topologies and sensor device architectures.展开更多
This paper focuses on the leader-following positive consensus problems of heterogeneous switched multi-agent systems.First,a state-feedback controller with dynamic compensation is introduced to achieve positive consen...This paper focuses on the leader-following positive consensus problems of heterogeneous switched multi-agent systems.First,a state-feedback controller with dynamic compensation is introduced to achieve positive consensus under average dwell time switching.Then sufficient conditions are derived to guarantee the positive consensus.The gain matrices of the control protocol are described using a matrix decomposition approach and the corresponding computational complexity is reduced by resorting to linear programming and co-positive Lyapunov functions.Finally,two numerical examples are provided to illustrate the results obtained.展开更多
Despite demonstrating significant anti-tumor potential as an artemisinin derivative,artesunate faces delivery efficiency challenges due to low water solubility and insufficient targeting specificity.To improve the del...Despite demonstrating significant anti-tumor potential as an artemisinin derivative,artesunate faces delivery efficiency challenges due to low water solubility and insufficient targeting specificity.To improve the delivery efficiency,we engineered three artesunate(ART) derivatives,AC_(15)-L(linear),AC_(15)-B(branched),and AC_(15)-C(cyclic) with distinct aliphatic chain architectures.Unexpectedly,we observed that AC_(15)-C exhibited superior cytotoxicity against 4T1 breast cancer cells,and had the highest binding affinity for Lon protease 1(LONP1)(-72.6 kcal/mol).Subsequently,disulfide bond-containing lipid-PEG(DSPESS-PEG2K) modified chain architecture-engineered ART derivatives nanoassemblies(NAs) were developed to mitigate solubility-related limitations while enhancing targeting precision.Molecular docking and experimental validation demonstrated that ART derivatives inhibited LONP1 through hydrophobic interactions while preserved Fe^(2+)-mediated Fenton-like reaction activity.In vitro and in vivo evaluations demonstrated that AC_(15)-C NAs outperformed free ART and other NAs,suppressing 4T1 tumor growth via dual action:LONP1-directed mitochondrial proteostasis collapse and reactive oxygen species(ROS) amplification through Fe^(2+)-ART interactions.This study elucidated a novel anti-tumor mechanism of ART through the rational design of derivatives with spatially configured aliphatic chains,and developed reductionresponsive NAs to provide an advanced delivery strategy.展开更多
With the advent of sixth-generation mobile communications(6G),space-air-ground integrated networks have become mainstream.This paper focuses on collaborative scheduling for mobile edge computing(MEC)under a three-tier...With the advent of sixth-generation mobile communications(6G),space-air-ground integrated networks have become mainstream.This paper focuses on collaborative scheduling for mobile edge computing(MEC)under a three-tier heterogeneous architecture composed of mobile devices,unmanned aerial vehicles(UAVs),and macro base stations(BSs).This scenario typically faces fast channel fading,dynamic computational loads,and energy constraints,whereas classical queuing-theoretic or convex-optimization approaches struggle to yield robust solutions in highly dynamic settings.To address this issue,we formulate a multi-agent Markov decision process(MDP)for an air-ground-fused MEC system,unify link selection,bandwidth/power allocation,and task offloading into a continuous action space and propose a joint scheduling strategy that is based on an improved MATD3 algorithm.The improvements include Alternating Layer Normalization(ALN)in the actor to suppress gradient variance,Residual Orthogonalization(RO)in the critic to reduce the correlation between the twin Q-value estimates,and a dynamic-temperature reward to enable adaptive trade-offs during training.On a multi-user,dual-link simulation platform,we conduct ablation and baseline comparisons.The results reveal that the proposed method has better convergence and stability.Compared with MADDPG,TD3,and DSAC,our algorithm achieves more robust performance across key metrics.展开更多
This paper investigates the consensus tracking control problem for high order nonlinear multi-agent systems subject to non-affine faults,partial measurable states,uncertain control coefficients,and unknown external di...This paper investigates the consensus tracking control problem for high order nonlinear multi-agent systems subject to non-affine faults,partial measurable states,uncertain control coefficients,and unknown external disturbances.Under the directed topology conditions,an observer-based finite-time control strategy based on adaptive backstepping and is proposed,in which a neural network-based state observer is employed to approximate the unmeasurable system state variables.To address the complexity explosion problem associated with the backstepping method,a finite-time command filter is incorporated,with error compensation signals designed to mitigate the filter-induced errors.Additionally,the Butterworth low-pass filter is introduced to avoid the algebraic ring problem in the design of the controller.The finite-time stability of the closed-loop system is rigorously analyzed with the finite-time Lyapunov stability criterion,validating that all closed-loop signals of the system remain bounded within a finite time.Finally,the effectiveness of the proposed control strategy is verified through a simulation example.展开更多
Deep learning has become integral to robotics,particularly in tasks such as robotic grasping,where objects often exhibit diverse shapes,textures,and physical properties.In robotic grasping tasks,due to the diverse cha...Deep learning has become integral to robotics,particularly in tasks such as robotic grasping,where objects often exhibit diverse shapes,textures,and physical properties.In robotic grasping tasks,due to the diverse characteristics of the targets,frequent adjustments to the network architecture and parameters are required to avoid a decrease in model accuracy,which presents a significant challenge for non-experts.Neural Architecture Search(NAS)provides a compelling method through the automated generation of network architectures,enabling the discovery of models that achieve high accuracy through efficient search algorithms.Compared to manually designed networks,NAS methods can significantly reduce design costs,time expenditure,and improve model performance.However,such methods often involve complex topological connections,and these redundant structures can severely reduce computational efficiency.To overcome this challenge,this work puts forward a robotic grasp detection framework founded on NAS.The method automatically designs a lightweight network with high accuracy and low topological complexity,effectively adapting to the target object to generate the optimal grasp pose,thereby significantly improving the success rate of robotic grasping.Additionally,we use Class Activation Mapping(CAM)as an interpretability tool,which captures sensitive information during the perception process through visualized results.The searched model achieved competitive,and in some cases superior,performance on the Cornell and Jacquard public datasets,achieving accuracies of 98.3%and 96.8%,respectively,while sustaining a detection speed of 89 frames per second with only 0.41 million parameters.To further validate its effectiveness beyond benchmark evaluations,we conducted real-world grasping experiments on a UR5 robotic arm,where the model demonstrated reliable performance across diverse objects and high grasp success rates,thereby confirming its practical applicability in robotic manipulation tasks.展开更多
Fault-tolerant systems are crucial for ensuring the reliability and availability of missioncritical applications in modern computing environments.The dynamic heterogeneous redundancy(DHR)architecture is a key componen...Fault-tolerant systems are crucial for ensuring the reliability and availability of missioncritical applications in modern computing environments.The dynamic heterogeneous redundancy(DHR)architecture is a key component in constructing fault-tolerant systems,particularly in areas such as national security,power networks,and banking private networks.DHR is transforming the cyberspace security industry chain by accommodating a broader range of applications and increasingly capturing the market.However,the development of applications for DHR architecture encounters challenges due to the complexities of handling heterogeneity,managing dynamism,and maintaining usability.To address these issues,we introduce MimicStudio,a comprehensive development framework with a standardized workflow.To our knowledge,MimicStudio is the first effective solution for DHR software development.We present a detailed implementation of MimicStudio with a heterogeneous microcontroller unit project,encompassing three CPUs with different instruction set architectures.The paper evaluates MimicStudio’s support for essential features,including zero-copy synchronization,parallelized build,multi-core collaborative debugging,and dynamic adjustment of the software system’s structure.Our results show that MimicStudio provides a flexible and efficient solution for supporting the dynamic,heterogeneous,and redundant features of fault-tolerant systems.展开更多
Multimodal dialogue systems often fail to maintain coherent reasoning over extended conversations and suffer from hallucination due to limited context modeling capabilities.Current approaches struggle with crossmodal ...Multimodal dialogue systems often fail to maintain coherent reasoning over extended conversations and suffer from hallucination due to limited context modeling capabilities.Current approaches struggle with crossmodal alignment,temporal consistency,and robust handling of noisy or incomplete inputs across multiple modalities.We propose Multi Agent-Chain of Thought(CoT),a novel multi-agent chain-of-thought reasoning framework where specialized agents for text,vision,and speech modalities collaboratively construct shared reasoning traces through inter-agent message passing and consensus voting mechanisms.Our architecture incorporates self-reflection modules,conflict resolution protocols,and dynamic rationale alignment to enhance consistency,factual accuracy,and user engagement.The framework employs a hierarchical attention mechanism with cross-modal fusion and implements adaptive reasoning depth based on dialogue complexity.Comprehensive evaluations on Situated Interactive Multi-Modal Conversations(SIMMC)2.0,VisDial v1.0,and newly introduced challenging scenarios demonstrate statistically significant improvements in grounding accuracy(p<0.01),chain-of-thought interpretability,and robustness to adversarial inputs compared to state-of-the-art monolithic transformer baselines and existing multi-agent approaches.展开更多
Improved yield potential is the goal of barley domestication and cultivation.During this process,two-and six-rowed barley types emerged and have been utilised in breeding and production.The six-rowed type could produc...Improved yield potential is the goal of barley domestication and cultivation.During this process,two-and six-rowed barley types emerged and have been utilised in breeding and production.The six-rowed type could produce three times as many grains as its ancestral two-rowed forms,thus dominating barley cultivation for thousands of years.The deficiens form of the two-rowed type,characterised by extremely suppressed lateral spikelets,has gained dominance over the past few decades in barley-growing regions worldwide.We hypothesised that the absence of lateral spikelets in deficiens barley affects spike architecture and spike-related traits,contributing to its superior yield potential of deficiens barley cultivation.Currently,a deficiens barley variety,RGT Planet,is the most popular barley variety in the world.In this study,we used two F_(2) populations derived from crossing RGT Planet with two canonical two-rowed barley and identified the functional allele Vrs1.t1 associated with deficiens morphology.We observed that the Vrs1.t1 allele may contribute to high yield potential by optimising spike architecture through increased spikelet length,grain number,and grain size.Phylogenetic analysis suggests that the deficiens mutation was likely present from the early stages of barley cultivation in the Fertile Crescent and spread to Ethiopia and beyond with agricultural expansion.We conclude that the ancient deficiens allele Vrs1.t1 has been a critical driver for the recent success of modern barley improvement by optimising spike architecture.展开更多
During the past years, a number of smart manufacturing concepts have been proposed, such as cloud manufacturing, Industry 4.0, and Industrial Internet. One of their common aims is to optimize the collaborative resourc...During the past years, a number of smart manufacturing concepts have been proposed, such as cloud manufacturing, Industry 4.0, and Industrial Internet. One of their common aims is to optimize the collaborative resource configuration across enterprises by establishing platforms that aggregate distributed resources. In all of these concepts, a complete manufacturing system consists of distributed physical manufacturing systems and a platform containing the virtual manufacturing systems mapped from the physical ones. We call such manufacturing systems platform-based smart manufacturing systems (PSMSs). A PSMS can therefore be regarded as a huge cyber-physical system with the cyber part being the platform and the physical part being the corresponding physical manufacturing system. A significant issue for a PSMS is how to optimally schedule the aggregated resources. Multi-agent technology provides an effective approach for solving this issue. In this paper we propose a multi-agent architecture for scheduling in PSMSs, which consists of a platform-level scheduling multi-agent system (MAS) and an enterprise-level scheduling MAS. Procedures, characteristics, and requirements of scheduling in PSMSs are presented. A model for sched-uling in a PSMS based on the architecture is proposed. A case study is conducted to demonstrate the effectiveness of the proposed architecture and model.展开更多
Traditional ERP software system cannot efficiently su pport new management ideas such as BPR, DEM and virtual enterprise which emphasi zes that enterprise should be adjusted to market changes and business process ch a...Traditional ERP software system cannot efficiently su pport new management ideas such as BPR, DEM and virtual enterprise which emphasi zes that enterprise should be adjusted to market changes and business process ch ain and value chain should be integrated tightly. To solve these problems, this paper proposed the conception of Flexible ERP system. F-ERP is a self- adapti ve software system based on multi-agent technology. It developed the followin g kind of agents which are useful for F-ERP: business process agent, interf ace agent, data agent and decision and analysis agent. The F-ERP software syste m is an hierarchy system which is composed of data layer, system tools layer, bu siness application layer and business decision layer. It used component based de velopment mythology and complied with CORBA to development F-ERP. The F-ERP sy stem can support the new management ideas such as BPR, DEM and virtual enterpris e etc. By implementation of it, enterprise can improve its management and promot e its competence.展开更多
World Wide Web (WWW) is a vast repository of information, including a great deal of geographic information. But the location and retrieval of geographic information will require a significant amount of time and effort...World Wide Web (WWW) is a vast repository of information, including a great deal of geographic information. But the location and retrieval of geographic information will require a significant amount of time and effort. In addition, different users usually have different views and interests in the same information. To resolve such problems, this paper first proposed a model of geographic information gathering based on multi-Agent (MA) architecture. Then based on this model, we construct a prototype system with GML (Geography Markup Language). This system consists of three tiers-Client, Web Server and Data Resource. Finally, we expatiate on the process of Web Server.展开更多
With the rapid development of flexible wearable electronics,the demand for stretchable energy storage devices has surged.In this work,a novel gradient-layered architecture was design based on single-pore hollow lignin...With the rapid development of flexible wearable electronics,the demand for stretchable energy storage devices has surged.In this work,a novel gradient-layered architecture was design based on single-pore hollow lignin nanospheres(HLNPs)-intercalated two-dimensional transition metal carbide(Ti_(3)C_(2)T_(x) MXene)for fabricating highly stretchable and durable supercapacitors.By depositing and inserting HLNPs in the MXene layers with a bottom-up decreasing gradient,a multilayered porous MXene structure with smooth ion channels was constructed by reducing the overstacking of MXene lamella.Moreover,the micro-chamber architecture of thin-walled lignin nanospheres effectively extended the contact area between lignin and MXene to improve ion and electron accessibility,thus better utilizing the pseudocapacitive property of lignin.All these strategies effectively enhanced the capacitive performance of the electrodes.In addition,HLNPs,which acted as a protective phase for MXene layer,enhanced mechanical properties of the wrinkled stretchable electrodes by releasing stress through slip and deformation during the stretch-release cycling and greatly improved the structural integrity and capacitive stability of the electrodes.Flexible electrodes and symmetric flexible all-solid-state supercapacitors capable of enduring 600%uniaxial tensile strain were developed with high specific capacitances of 1273 mF cm^(−2)(241 F g^(−1))and 514 mF cm^(−2)(95 F g^(−1)),respectively.Moreover,their capacitances were well preserved after 1000 times of 600%stretch-release cycling.This study showcased new possibilities of incorporating biobased lignin nanospheres in energy storage devices to fabricate stretchable devices leveraging synergies among various two-dimensional nanomaterials.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
Multi-agent systems(MASs)have demonstrated significant achievements in a wide range of tasks,leveraging their capacity for coordination and adaptation within complex environments.Moreover,the enhancement of their inte...Multi-agent systems(MASs)have demonstrated significant achievements in a wide range of tasks,leveraging their capacity for coordination and adaptation within complex environments.Moreover,the enhancement of their intelligent functionalities is crucial for tackling increasingly challenging tasks.This goal resonates with a paradigm shift within the artificial intelligence(AI)community,from“internet AI”to“embodied AI”,and the MASs with embodied AI are referred to as embodied multi-agent systems(EMASs).An EMAS has the potential to acquire generalized competencies through interactions with environments,enabling it to effectively address a variety of tasks and thereby make a substantial contribution to the quest for artificial general intelligence.Despite the burgeoning interest in this domain,a comprehensive review of EMAS has been lacking.This paper offers analysis and synthesis for EMASs from a control perspective,conceptualizing each embodied agent as an entity equipped with a“brain”for decision and a“body”for environmental interaction.System designs are classified into open-loop,closed-loop,and double-loop categories,and EMAS implementations are discussed.Additionally,the current applications and challenges faced by EMASs are summarized and potential avenues for future research in this field are provided.展开更多
High-density planting increases maize yield but also canopy crowding and stalk lodging.Aiming this contradiction,a wavy canopy was created using interlaced chemical application(IC)of a plant growth retardant at the V1...High-density planting increases maize yield but also canopy crowding and stalk lodging.Aiming this contradiction,a wavy canopy was created using interlaced chemical application(IC)of a plant growth retardant at the V14 stage with three densities(60,000,75,000,and 90,000 plants ha-1,indicated by D1,D2,and D3,respectively)for two seasons.The results showed that the IC-treated wavy canopy featuring both natural height(IC-H)and dwarfed(IC-L)plants,improved light transmission by 8.54%,8.49%,and 16.49%on average than the corresponding controls(CK)at D1,D2,and D3,respectively.The alleviation of canopy crowding stimulated leaf photosynthesis,sugar availability,basal-internode strength,and decreased plant lodging ratios in both IC-H and IC-L,particularly under higher densities.Meanwhile,the IC populations produced significantly higher yield than CK,with an average increase of 3.38%,16.70%,and 15.28%at D1,D2,and D3,respectively.Collectively,this study proposed a new wavy canopy strategy using plant growth retardant to simultaneously increase yield performance and lodging resistance,thus offering a sustainable solution for further development of high-density maize production.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 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
基金This research was funded by the Project of the National Natural Science Foundation of China,Grant Number 62106283.
文摘Aiming at the problems of low solution accuracy and high decision pressure when facing large-scale dynamic task allocation(DTA)and high-dimensional decision space with single agent,this paper combines the deep reinforce-ment learning(DRL)theory and an improved Multi-Agent Deep Deterministic Policy Gradient(MADDPG-D2)algorithm with a dual experience replay pool and a dual noise based on multi-agent architecture is proposed to improve the efficiency of DTA.The algorithm is based on the traditional Multi-Agent Deep Deterministic Policy Gradient(MADDPG)algorithm,and considers the introduction of a double noise mechanism to increase the action exploration space in the early stage of the algorithm,and the introduction of a double experience pool to improve the data utilization rate;at the same time,in order to accelerate the training speed and efficiency of the agents,and to solve the cold-start problem of the training,the a priori knowledge technology is applied to the training of the algorithm.Finally,the MADDPG-D2 algorithm is compared and analyzed based on the digital battlefield of ground and air confrontation.The experimental results show that the agents trained by the MADDPG-D2 algorithm have higher win rates and average rewards,can utilize the resources more reasonably,and better solve the problem of the traditional single agent algorithms facing the difficulty of solving the problem in the high-dimensional decision space.The MADDPG-D2 algorithm based on multi-agent architecture proposed in this paper has certain superiority and rationality in DTA.
文摘Complex and distributed systems are more and more associated with the application of WSN (Wireless Sensor Network) technology. The design of such applications presents important challenges and requires the assistance of several methodologies and tools. Multi-Agent systems (MAS) have been identified as one of the most suitable technologies to contribute to this domain due to their appropriateness for modeling distributed and autonomous complex systems. This work aims to contribute in the help of the design of WSN applications. The proposed architecture exploits the advantages of MAS for modeling WSN services, network topologies and sensor device architectures.
基金supported by the National Natural Science Foundation of China(62463007,62463005)the Natural Science Foundation of Hainan Province(625RC710,625MS047)+1 种基金the System Control and Information Processing Education Ministry Key Laboratory Open Funding,China(Scip20240119)the Science Research Funding of Hainan University,China(KYQD(ZR)22180,KYQD(ZR)23180).
文摘This paper focuses on the leader-following positive consensus problems of heterogeneous switched multi-agent systems.First,a state-feedback controller with dynamic compensation is introduced to achieve positive consensus under average dwell time switching.Then sufficient conditions are derived to guarantee the positive consensus.The gain matrices of the control protocol are described using a matrix decomposition approach and the corresponding computational complexity is reduced by resorting to linear programming and co-positive Lyapunov functions.Finally,two numerical examples are provided to illustrate the results obtained.
基金financially supported by the Liaoning Revitalization Talents Program (No.XLYC2403107)the Excellent Youth Science Foundation of Liaoning Province (No.2024JH3/10200046)the Basic Scientific Research Project of Liaoning Provincial Department of Education (No.LJ212410163015)。
文摘Despite demonstrating significant anti-tumor potential as an artemisinin derivative,artesunate faces delivery efficiency challenges due to low water solubility and insufficient targeting specificity.To improve the delivery efficiency,we engineered three artesunate(ART) derivatives,AC_(15)-L(linear),AC_(15)-B(branched),and AC_(15)-C(cyclic) with distinct aliphatic chain architectures.Unexpectedly,we observed that AC_(15)-C exhibited superior cytotoxicity against 4T1 breast cancer cells,and had the highest binding affinity for Lon protease 1(LONP1)(-72.6 kcal/mol).Subsequently,disulfide bond-containing lipid-PEG(DSPESS-PEG2K) modified chain architecture-engineered ART derivatives nanoassemblies(NAs) were developed to mitigate solubility-related limitations while enhancing targeting precision.Molecular docking and experimental validation demonstrated that ART derivatives inhibited LONP1 through hydrophobic interactions while preserved Fe^(2+)-mediated Fenton-like reaction activity.In vitro and in vivo evaluations demonstrated that AC_(15)-C NAs outperformed free ART and other NAs,suppressing 4T1 tumor growth via dual action:LONP1-directed mitochondrial proteostasis collapse and reactive oxygen species(ROS) amplification through Fe^(2+)-ART interactions.This study elucidated a novel anti-tumor mechanism of ART through the rational design of derivatives with spatially configured aliphatic chains,and developed reductionresponsive NAs to provide an advanced delivery strategy.
文摘With the advent of sixth-generation mobile communications(6G),space-air-ground integrated networks have become mainstream.This paper focuses on collaborative scheduling for mobile edge computing(MEC)under a three-tier heterogeneous architecture composed of mobile devices,unmanned aerial vehicles(UAVs),and macro base stations(BSs).This scenario typically faces fast channel fading,dynamic computational loads,and energy constraints,whereas classical queuing-theoretic or convex-optimization approaches struggle to yield robust solutions in highly dynamic settings.To address this issue,we formulate a multi-agent Markov decision process(MDP)for an air-ground-fused MEC system,unify link selection,bandwidth/power allocation,and task offloading into a continuous action space and propose a joint scheduling strategy that is based on an improved MATD3 algorithm.The improvements include Alternating Layer Normalization(ALN)in the actor to suppress gradient variance,Residual Orthogonalization(RO)in the critic to reduce the correlation between the twin Q-value estimates,and a dynamic-temperature reward to enable adaptive trade-offs during training.On a multi-user,dual-link simulation platform,we conduct ablation and baseline comparisons.The results reveal that the proposed method has better convergence and stability.Compared with MADDPG,TD3,and DSAC,our algorithm achieves more robust performance across key metrics.
基金supported in part by the Beijing Natural Science Foundation under Grant 4252050in part by the National Science Fund for Distinguished Young Scholars under Grant 62425304in part by the Basic Science Center Programs of NSFC under Grant 62088101.
文摘This paper investigates the consensus tracking control problem for high order nonlinear multi-agent systems subject to non-affine faults,partial measurable states,uncertain control coefficients,and unknown external disturbances.Under the directed topology conditions,an observer-based finite-time control strategy based on adaptive backstepping and is proposed,in which a neural network-based state observer is employed to approximate the unmeasurable system state variables.To address the complexity explosion problem associated with the backstepping method,a finite-time command filter is incorporated,with error compensation signals designed to mitigate the filter-induced errors.Additionally,the Butterworth low-pass filter is introduced to avoid the algebraic ring problem in the design of the controller.The finite-time stability of the closed-loop system is rigorously analyzed with the finite-time Lyapunov stability criterion,validating that all closed-loop signals of the system remain bounded within a finite time.Finally,the effectiveness of the proposed control strategy is verified through a simulation example.
基金funded by Guangdong Basic and Applied Basic Research Foundation(2023B1515120064)National Natural Science Foundation of China(62273097).
文摘Deep learning has become integral to robotics,particularly in tasks such as robotic grasping,where objects often exhibit diverse shapes,textures,and physical properties.In robotic grasping tasks,due to the diverse characteristics of the targets,frequent adjustments to the network architecture and parameters are required to avoid a decrease in model accuracy,which presents a significant challenge for non-experts.Neural Architecture Search(NAS)provides a compelling method through the automated generation of network architectures,enabling the discovery of models that achieve high accuracy through efficient search algorithms.Compared to manually designed networks,NAS methods can significantly reduce design costs,time expenditure,and improve model performance.However,such methods often involve complex topological connections,and these redundant structures can severely reduce computational efficiency.To overcome this challenge,this work puts forward a robotic grasp detection framework founded on NAS.The method automatically designs a lightweight network with high accuracy and low topological complexity,effectively adapting to the target object to generate the optimal grasp pose,thereby significantly improving the success rate of robotic grasping.Additionally,we use Class Activation Mapping(CAM)as an interpretability tool,which captures sensitive information during the perception process through visualized results.The searched model achieved competitive,and in some cases superior,performance on the Cornell and Jacquard public datasets,achieving accuracies of 98.3%and 96.8%,respectively,while sustaining a detection speed of 89 frames per second with only 0.41 million parameters.To further validate its effectiveness beyond benchmark evaluations,we conducted real-world grasping experiments on a UR5 robotic arm,where the model demonstrated reliable performance across diverse objects and high grasp success rates,thereby confirming its practical applicability in robotic manipulation tasks.
基金supported by National Key Research and Development Program of China(No.2023YFB 4404200).
文摘Fault-tolerant systems are crucial for ensuring the reliability and availability of missioncritical applications in modern computing environments.The dynamic heterogeneous redundancy(DHR)architecture is a key component in constructing fault-tolerant systems,particularly in areas such as national security,power networks,and banking private networks.DHR is transforming the cyberspace security industry chain by accommodating a broader range of applications and increasingly capturing the market.However,the development of applications for DHR architecture encounters challenges due to the complexities of handling heterogeneity,managing dynamism,and maintaining usability.To address these issues,we introduce MimicStudio,a comprehensive development framework with a standardized workflow.To our knowledge,MimicStudio is the first effective solution for DHR software development.We present a detailed implementation of MimicStudio with a heterogeneous microcontroller unit project,encompassing three CPUs with different instruction set architectures.The paper evaluates MimicStudio’s support for essential features,including zero-copy synchronization,parallelized build,multi-core collaborative debugging,and dynamic adjustment of the software system’s structure.Our results show that MimicStudio provides a flexible and efficient solution for supporting the dynamic,heterogeneous,and redundant features of fault-tolerant systems.
文摘Multimodal dialogue systems often fail to maintain coherent reasoning over extended conversations and suffer from hallucination due to limited context modeling capabilities.Current approaches struggle with crossmodal alignment,temporal consistency,and robust handling of noisy or incomplete inputs across multiple modalities.We propose Multi Agent-Chain of Thought(CoT),a novel multi-agent chain-of-thought reasoning framework where specialized agents for text,vision,and speech modalities collaboratively construct shared reasoning traces through inter-agent message passing and consensus voting mechanisms.Our architecture incorporates self-reflection modules,conflict resolution protocols,and dynamic rationale alignment to enhance consistency,factual accuracy,and user engagement.The framework employs a hierarchical attention mechanism with cross-modal fusion and implements adaptive reasoning depth based on dialogue complexity.Comprehensive evaluations on Situated Interactive Multi-Modal Conversations(SIMMC)2.0,VisDial v1.0,and newly introduced challenging scenarios demonstrate statistically significant improvements in grounding accuracy(p<0.01),chain-of-thought interpretability,and robustness to adversarial inputs compared to state-of-the-art monolithic transformer baselines and existing multi-agent approaches.
基金Funding for this research was provided by the Australia Grain Research and Development Corporation(9176507)the Western Crop Genetics Alliance.Jingye Cheng thanks The University of Tasmania,Australia for the scholarship(495802)。
文摘Improved yield potential is the goal of barley domestication and cultivation.During this process,two-and six-rowed barley types emerged and have been utilised in breeding and production.The six-rowed type could produce three times as many grains as its ancestral two-rowed forms,thus dominating barley cultivation for thousands of years.The deficiens form of the two-rowed type,characterised by extremely suppressed lateral spikelets,has gained dominance over the past few decades in barley-growing regions worldwide.We hypothesised that the absence of lateral spikelets in deficiens barley affects spike architecture and spike-related traits,contributing to its superior yield potential of deficiens barley cultivation.Currently,a deficiens barley variety,RGT Planet,is the most popular barley variety in the world.In this study,we used two F_(2) populations derived from crossing RGT Planet with two canonical two-rowed barley and identified the functional allele Vrs1.t1 associated with deficiens morphology.We observed that the Vrs1.t1 allele may contribute to high yield potential by optimising spike architecture through increased spikelet length,grain number,and grain size.Phylogenetic analysis suggests that the deficiens mutation was likely present from the early stages of barley cultivation in the Fertile Crescent and spread to Ethiopia and beyond with agricultural expansion.We conclude that the ancient deficiens allele Vrs1.t1 has been a critical driver for the recent success of modern barley improvement by optimising spike architecture.
基金Project supported by the National Natural Science Foundation of China(Nos.61973243,61873014,and 51875030)the National Key Research and Development Program of China(No.2018YFB1702703)。
文摘During the past years, a number of smart manufacturing concepts have been proposed, such as cloud manufacturing, Industry 4.0, and Industrial Internet. One of their common aims is to optimize the collaborative resource configuration across enterprises by establishing platforms that aggregate distributed resources. In all of these concepts, a complete manufacturing system consists of distributed physical manufacturing systems and a platform containing the virtual manufacturing systems mapped from the physical ones. We call such manufacturing systems platform-based smart manufacturing systems (PSMSs). A PSMS can therefore be regarded as a huge cyber-physical system with the cyber part being the platform and the physical part being the corresponding physical manufacturing system. A significant issue for a PSMS is how to optimally schedule the aggregated resources. Multi-agent technology provides an effective approach for solving this issue. In this paper we propose a multi-agent architecture for scheduling in PSMSs, which consists of a platform-level scheduling multi-agent system (MAS) and an enterprise-level scheduling MAS. Procedures, characteristics, and requirements of scheduling in PSMSs are presented. A model for sched-uling in a PSMS based on the architecture is proposed. A case study is conducted to demonstrate the effectiveness of the proposed architecture and model.
文摘Traditional ERP software system cannot efficiently su pport new management ideas such as BPR, DEM and virtual enterprise which emphasi zes that enterprise should be adjusted to market changes and business process ch ain and value chain should be integrated tightly. To solve these problems, this paper proposed the conception of Flexible ERP system. F-ERP is a self- adapti ve software system based on multi-agent technology. It developed the followin g kind of agents which are useful for F-ERP: business process agent, interf ace agent, data agent and decision and analysis agent. The F-ERP software syste m is an hierarchy system which is composed of data layer, system tools layer, bu siness application layer and business decision layer. It used component based de velopment mythology and complied with CORBA to development F-ERP. The F-ERP sy stem can support the new management ideas such as BPR, DEM and virtual enterpris e etc. By implementation of it, enterprise can improve its management and promot e its competence.
文摘World Wide Web (WWW) is a vast repository of information, including a great deal of geographic information. But the location and retrieval of geographic information will require a significant amount of time and effort. In addition, different users usually have different views and interests in the same information. To resolve such problems, this paper first proposed a model of geographic information gathering based on multi-Agent (MA) architecture. Then based on this model, we construct a prototype system with GML (Geography Markup Language). This system consists of three tiers-Client, Web Server and Data Resource. Finally, we expatiate on the process of Web Server.
基金supported by Natural Science and Engineering Research Council of Canada(RGPIN-2017-06737)Canada Research Chairs program,the National Key Research and Development Program of China(2017YFD0601005,2022YFD0904201)+1 种基金the National Natural Science Foundation of China(51203075)the China Scholarship Council(Grant No.CSC202208320361).
文摘With the rapid development of flexible wearable electronics,the demand for stretchable energy storage devices has surged.In this work,a novel gradient-layered architecture was design based on single-pore hollow lignin nanospheres(HLNPs)-intercalated two-dimensional transition metal carbide(Ti_(3)C_(2)T_(x) MXene)for fabricating highly stretchable and durable supercapacitors.By depositing and inserting HLNPs in the MXene layers with a bottom-up decreasing gradient,a multilayered porous MXene structure with smooth ion channels was constructed by reducing the overstacking of MXene lamella.Moreover,the micro-chamber architecture of thin-walled lignin nanospheres effectively extended the contact area between lignin and MXene to improve ion and electron accessibility,thus better utilizing the pseudocapacitive property of lignin.All these strategies effectively enhanced the capacitive performance of the electrodes.In addition,HLNPs,which acted as a protective phase for MXene layer,enhanced mechanical properties of the wrinkled stretchable electrodes by releasing stress through slip and deformation during the stretch-release cycling and greatly improved the structural integrity and capacitive stability of the electrodes.Flexible electrodes and symmetric flexible all-solid-state supercapacitors capable of enduring 600%uniaxial tensile strain were developed with high specific capacitances of 1273 mF cm^(−2)(241 F g^(−1))and 514 mF cm^(−2)(95 F g^(−1)),respectively.Moreover,their capacitances were well preserved after 1000 times of 600%stretch-release cycling.This study showcased new possibilities of incorporating biobased lignin nanospheres in energy storage devices to fabricate stretchable devices leveraging synergies among various two-dimensional nanomaterials.
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.
基金supported in part by National Natural Science Foundation of China(62495095,62088101).
文摘Multi-agent systems(MASs)have demonstrated significant achievements in a wide range of tasks,leveraging their capacity for coordination and adaptation within complex environments.Moreover,the enhancement of their intelligent functionalities is crucial for tackling increasingly challenging tasks.This goal resonates with a paradigm shift within the artificial intelligence(AI)community,from“internet AI”to“embodied AI”,and the MASs with embodied AI are referred to as embodied multi-agent systems(EMASs).An EMAS has the potential to acquire generalized competencies through interactions with environments,enabling it to effectively address a variety of tasks and thereby make a substantial contribution to the quest for artificial general intelligence.Despite the burgeoning interest in this domain,a comprehensive review of EMAS has been lacking.This paper offers analysis and synthesis for EMASs from a control perspective,conceptualizing each embodied agent as an entity equipped with a“brain”for decision and a“body”for environmental interaction.System designs are classified into open-loop,closed-loop,and double-loop categories,and EMAS implementations are discussed.Additionally,the current applications and challenges faced by EMASs are summarized and potential avenues for future research in this field are provided.
基金supported by the National Key Research and Development Program of China(2023YFD2303302,2022YFD2300803)the National Natural Science Foundation of China(32160445)the China Agriculture Research System of MOF and MARA(CARS-02-16).
文摘High-density planting increases maize yield but also canopy crowding and stalk lodging.Aiming this contradiction,a wavy canopy was created using interlaced chemical application(IC)of a plant growth retardant at the V14 stage with three densities(60,000,75,000,and 90,000 plants ha-1,indicated by D1,D2,and D3,respectively)for two seasons.The results showed that the IC-treated wavy canopy featuring both natural height(IC-H)and dwarfed(IC-L)plants,improved light transmission by 8.54%,8.49%,and 16.49%on average than the corresponding controls(CK)at D1,D2,and D3,respectively.The alleviation of canopy crowding stimulated leaf photosynthesis,sugar availability,basal-internode strength,and decreased plant lodging ratios in both IC-H and IC-L,particularly under higher densities.Meanwhile,the IC populations produced significantly higher yield than CK,with an average increase of 3.38%,16.70%,and 15.28%at D1,D2,and D3,respectively.Collectively,this study proposed a new wavy canopy strategy using plant growth retardant to simultaneously increase yield performance and lodging resistance,thus offering a sustainable solution for further development of high-density maize production.
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.
