Multi-agent decision support system for missile defense based on improved PSO algorithm

Ballistic missile defense system (BMDS) is important for its special role in ensuring national security and maintaining strategic balance. Research on modeling and simulation of the BMDS beforehand is essential as developing a real one requires lots of manpower and resources. BMDS is a typical complex system for its nonlinear, adaptive and uncertainty characteristics. The agent-based modeling method is well suited for the complex system whose overall behaviors are determined by interactions among individual elements. A multi-agent decision support system (DSS), which includes missile agent, radar agent and command center agent, is established based on the studies of structure and function of BMDS. Considering the constraints brought by radar, intercept missile, offensive missile and commander, the objective function of DSS is established. In order to dynamically generate the optimal interception plan, the variable neighborhood negative selection particle swarm optimization (VNNSPSO) algorithm is proposed to support the decision making of DSS. The proposed algorithm is compared with the standard PSO, constriction factor PSO (CFPSO), inertia weight linear decrease PSO (LDPSO), variable neighborhood PSO (VNPSO) algorithm from the aspects of convergence rate, iteration number, average fitness value and standard deviation. The simulation results verify the efficiency of the proposed algorithm. The multi-agent DSS is developed through the Repast simulation platform and the constructed DSS can generate intercept plans automatically and support three-dimensional dynamic display of missile defense process.

Offline Pre-trained Multi-agent Decision Transformer

Offline reinforcement learning leverages previously collected offline datasets to learn optimal policies with no necessity to access the real environment.Such a paradigm is also desirable for multi-agent reinforcement learning(MARL)tasks,given the combinatorially increased interactions among agents and with the environment.However,in MARL,the paradigm of offline pre-training with online fine-tuning has not been studied,nor even datasets or benchmarks for offline MARL research are available.In this paper,we facilitate the research by providing large-scale datasets and using them to examine the usage of the decision transformer in the context of MARL.We investigate the generalization of MARL offline pre-training in the following three aspects:1)between single agents and multiple agents,2)from offline pretraining to online fine tuning,and 3)to that of multiple downstream tasks with few-shot and zero-shot capabilities.We start by introducing the first offline MARL dataset with diverse quality levels based on the StarCraftII environment,and then propose the novel architecture of multi-agent decision transformer(MADT)for effective offline learning.MADT leverages the transformer′s modelling ability for sequence modelling and integrates it seamlessly with both offline and online MARL tasks.A significant benefit of MADT is that it learns generalizable policies that can transfer between different types of agents under different task scenarios.On the StarCraft II offline dataset,MADT outperforms the state-of-the-art offline reinforcement learning(RL)baselines,including BCQ and CQL.When applied to online tasks,the pre-trained MADT significantly improves sample efficiency and enjoys strong performance in both few-short and zero-shot cases.To the best of our knowledge,this is the first work that studies and demonstrates the effectiveness of offline pre-trained models in terms of sample efficiency and generalizability enhancements for MARL.

A single-task and multi-decision evolutionary game model based on multi-agent reinforcement learning

In the evolutionary game of the same task for groups,the changes in game rules,personal interests,the crowd size,and external supervision cause uncertain effects on individual decision-making and game results.In the Markov decision framework,a single-task multi-decision evolutionary game model based on multi-agent reinforcement learning is proposed to explore the evolutionary rules in the process of a game.The model can improve the result of a evolutionary game and facilitate the completion of the task.First,based on the multi-agent theory,to solve the existing problems in the original model,a negative feedback tax penalty mechanism is proposed to guide the strategy selection of individuals in the group.In addition,in order to evaluate the evolutionary game results of the group in the model,a calculation method of the group intelligence level is defined.Secondly,the Q-learning algorithm is used to improve the guiding effect of the negative feedback tax penalty mechanism.In the model,the selection strategy of the Q-learning algorithm is improved and a bounded rationality evolutionary game strategy is proposed based on the rule of evolutionary games and the consideration of the bounded rationality of individuals.Finally,simulation results show that the proposed model can effectively guide individuals to choose cooperation strategies which are beneficial to task completion and stability under different negative feedback factor values and different group sizes,so as to improve the group intelligence level.

Discrete decision model and multi-agent simulation of the Liang Zong two-chain hierarchical organization in a complex project

Different from the organization structure of complex projects in Western countries, the Liang Zong hierarchical organization structure of complex projects in China has two different chains, the chief-engineer chain and the general-director chain,to handle the trade-off between technical and management decisions. However, previous works on organization search have mainly focused on the single-chain hierarchical organization in which all decisions are regarded as homogeneous. The heterogeneity and the interdependency between technical decisions and management decisions have been neglected. A two-chain hierarchical organization structure mapped from a real complex project is constructed. Then, a discrete decision model for a Liang Zong two-chain hierarchical organization in an NK model framework is proposed. This model proves that this kind of organization structure can reduce the search space by a large amount and that the search process should reach a final stable state more quickly. For a more complicated decision mechanism, a multi-agent simulation based on the above NK model is used to explore the effect of the two-chain organization structure on the speed, stability, and performance of the search process. The results provide three insights into how, compared with the single-chain hierarchical organization, the two-chain organization can improve the search process: it can reduce the number of iterations efficiently; the search is more stable because the search space is a smoother hill-like fitness landscape; in general, the search performance can be improved.However, when the organization structure is very complicated, the performance of a two-chain organization is inferior to that of a single-chain organization. These findings about the efficiency of the unique Chinese-style organization structure can be used to guide organization design for complex projects.

Decision-Making Model of Public Service Facility Configuration in Multi-agent Communities and Its Application

The decision.making process of the public service facility configuration in multi.agent community is usually simplistic and static. In order to reflect dynamic changes and interactions of all behavior subjects induding of residents, real estate developers and the government, a decision-making model of public service facility configuration according to the multi-agent theory was made to improve the efficiency of the public service facility configuration in community and the living quality of residents. Taking a community to the cast of Jinhui Port in Fengxian District in Shanghai for example, the model analyzed the decision-makers＇ adaptive behaviors and simulated the decision.making criteria. The results indicate that the decision-making model and criteria can be well of satisfying the purpose of improving validity and rationality of public service facility configuration in large community.

Multi-Agent Decision Making in Large-Scale Systems

In this,the decision problem in a large-scale system consisting of sevral subeys subeystm are considerde and the metheds for cofliect between the subsystems are explored.Based on the multiperson multiobjective conflict decision(MMCD) model proposed in Ref.[6],the concept,of bargaining solution for conflicts in large-scale systems is presented,and an approach to achieving th bargaining soltion is proposed.

A Decision Support System Based on Multi-Agent Technology for Gene Expression Analysis

The genetic microarrays give to researchers a huge amount of data of many diseases represented by intensities of gene expression. In genomic medicine gene expression analysis is guided to find strategies for prevention and treatment of diseases with high rate of mortality like the different cancers. So, genomic medicine requires the use of complex information technology. The purpose of our paper is to present a multi-agent system developed in order to improve gene expression analysis with the automation of tasks about identification of genes involved in a cancer, and classification of tumors according to molecular biology. Agents that integrate the system, carry out reading files of intensity data of genes from microarrays, pre-processing of this information, and with machine learning methods make groups of genes involved in the process of a disease as well as the classification of samples that could propose new subtypes of tumors difficult to identify based on their morphology. Our results we prove that the multi-agent system requires a minimal intervention of user, and the agents generate knowledge that reduce the time and complexity of the work of prevention and diagnosis, and thus allow a more effective treatment of tumors.

基于Multi-Agent的无人机集群体系自主作战系统设计

针对无人集群自主作战体系设计中的关键问题,提出基于Multi-Agent的无人集群自主作战系统设计方法。建立无人集群各节点的Agent模型及其推演规则;对于仿真系统模块化和通用化的需求,设计系统互操作式接口和无人集群自主作战的交互关系;开展无人集群系统仿真推演验证。仿真结果表明,所提设计方案不仅能够有效开展并完成自主作战网络生成-集群演化-效能评估的全过程动态演示验证,而且能够通过重复随机试验进一步评估无人集群的协同作战效能,最后总结了集群协同作战的策略和经验。

COLLISION AVOIDANCE DECISION-MAKING MODEL OF MULTI-AGENTS IN VIRTUAL DRIVING ENVIRONMENT WITH ANALYTIC HIERARCHY PROCESS

Collision avoidance decision-making models of multiple agents in virtual driving environment are studied. Based on the behavioral characteristics and hierarchical structure of the collision avoidance decision-making in real life driving, delphi approach and mathematical statistics method are introduced to construct pair-wise comparison judgment matrix of collision avoidance decision choices to each collision situation. Analytic hierarchy process （AHP） is adopted to establish the agents＇ collision avoidance decision-making model. To simulate drivers＇ characteristics, driver factors are added to categorize driving modes into impatient mode, normal mode, and the cautious mode. The results show that this model can simulate human＇s thinking process, and the agents in the virtual environment can deal with collision situations and make decisions to avoid collisions without intervention. The model can also reflect diversity and uncertainly of real life driving behaviors, and solves the multi-objective, multi-choice ranking priority problem in multi-vehicle collision scenarios. This collision avoidance model of multi-agents model is feasible and effective, and can provide richer and closer-to-life virtual scene for driving simulator, reflecting real-life traffic environment more truly, this model can also promote the practicality of driving simulator.

基于Multi-Agent的水电站变压器故障诊断系统

为了精准、快速完成水电站变压器的故障诊断,设计基于Multi-Agent的水电站变压器故障诊断系统。变压器状态监控agent将检测到的变压器故障信息发送给系统管理agent,系统管理agent通过通信agent将变压器故障信息发送给变压器故障诊断agent,变压器故障诊断agent利用小波变换方法提取变压器故障特征,并将其作为IFOA-SVM模型输入,完成变压器故障分类后,获取变压器故障诊断结果,该结果通过通信agent显示给用户。实验表明,该系统可有效诊断变压器故障诊断,诊断成功率受系统故障信息丢失率的影响较小,诊断耗时、耗能小,并具有较高故障诊断成功率。

Stress-assisted corrosion mechanism of 3Ni steel by using gradient boosting decision tree machining learning method

Traditional 3Ni weathering steel cannot completely meet the requirements for offshore engineering development,resulting in the design of novel 3Ni steel with the addition of microalloy elements such as Mn or Nb for strength enhancement becoming a trend.The stress-assisted corrosion behavior of a novel designed high-strength 3Ni steel was investigated in the current study using the corrosion big data method.The information on the corrosion process was recorded using the galvanic corrosion current monitoring method.The gradi-ent boosting decision tree(GBDT)machine learning method was used to mine the corrosion mechanism,and the importance of the struc-ture factor was investigated.Field exposure tests were conducted to verify the calculated results using the GBDT method.Results indic-ated that the GBDT method can be effectively used to study the influence of structural factors on the corrosion process of 3Ni steel.Dif-ferent mechanisms for the addition of Mn and Cu to the stress-assisted corrosion of 3Ni steel suggested that Mn and Cu have no obvious effect on the corrosion rate of non-stressed 3Ni steel during the early stage of corrosion.When the corrosion reached a stable state,the in-crease in Mn element content increased the corrosion rate of 3Ni steel,while Cu reduced this rate.In the presence of stress,the increase in Mn element content and Cu addition can inhibit the corrosion process.The corrosion law of outdoor-exposed 3Ni steel is consistent with the law based on corrosion big data technology,verifying the reliability of the big data evaluation method and data prediction model selection.

UAV-Assisted Dynamic Avatar Task Migration for Vehicular Metaverse Services: A Multi-Agent Deep Reinforcement Learning Approach

Avatars, as promising digital representations and service assistants of users in Metaverses, can enable drivers and passengers to immerse themselves in 3D virtual services and spaces of UAV-assisted vehicular Metaverses. However, avatar tasks include a multitude of human-to-avatar and avatar-to-avatar interactive applications, e.g., augmented reality navigation,which consumes intensive computing resources. It is inefficient and impractical for vehicles to process avatar tasks locally. Fortunately, migrating avatar tasks to the nearest roadside units(RSU)or unmanned aerial vehicles(UAV) for execution is a promising solution to decrease computation overhead and reduce task processing latency, while the high mobility of vehicles brings challenges for vehicles to independently perform avatar migration decisions depending on current and future vehicle status. To address these challenges, in this paper, we propose a novel avatar task migration system based on multi-agent deep reinforcement learning(MADRL) to execute immersive vehicular avatar tasks dynamically. Specifically, we first formulate the problem of avatar task migration from vehicles to RSUs/UAVs as a partially observable Markov decision process that can be solved by MADRL algorithms. We then design the multi-agent proximal policy optimization(MAPPO) approach as the MADRL algorithm for the avatar task migration problem. To overcome slow convergence resulting from the curse of dimensionality and non-stationary issues caused by shared parameters in MAPPO, we further propose a transformer-based MAPPO approach via sequential decision-making models for the efficient representation of relationships among agents. Finally, to motivate terrestrial or non-terrestrial edge servers(e.g., RSUs or UAVs) to share computation resources and ensure traceability of the sharing records, we apply smart contracts and blockchain technologies to achieve secure sharing management. Numerical results demonstrate that the proposed approach outperforms the MAPPO approach by around 2% and effectively reduces approximately 20% of the latency of avatar task execution in UAV-assisted vehicular Metaverses.

Finite-time Prescribed Performance Time-Varying Formation Control for Second-Order Multi-Agent Systems With Non-Strict Feedback Based on a Neural Network Observer

This paper studies the problem of time-varying formation control with finite-time prescribed performance for nonstrict feedback second-order multi-agent systems with unmeasured states and unknown nonlinearities.To eliminate nonlinearities,neural networks are applied to approximate the inherent dynamics of the system.In addition,due to the limitations of the actual working conditions,each follower agent can only obtain the locally measurable partial state information of the leader agent.To address this problem,a neural network state observer based on the leader state information is designed.Then,a finite-time prescribed performance adaptive output feedback control strategy is proposed by restricting the sliding mode surface to a prescribed region,which ensures that the closed-loop system has practical finite-time stability and that formation errors of the multi-agent systems converge to the prescribed performance bound in finite time.Finally,a numerical simulation is provided to demonstrate the practicality and effectiveness of the developed algorithm.

Cognitive interference decision method for air defense missile fuze based on reinforcement learning

To solve the problem of the low interference success rate of air defense missile radio fuzes due to the unified interference form of the traditional fuze interference system,an interference decision method based Q-learning algorithm is proposed.First,dividing the distance between the missile and the target into multiple states to increase the quantity of state spaces.Second,a multidimensional motion space is utilized,and the search range of which changes with the distance of the projectile,to select parameters and minimize the amount of ineffective interference parameters.The interference effect is determined by detecting whether the fuze signal disappears.Finally,a weighted reward function is used to determine the reward value based on the range state,output power,and parameter quantity information of the interference form.The effectiveness of the proposed method in selecting the range of motion space parameters and designing the discrimination degree of the reward function has been verified through offline experiments involving full-range missile rendezvous.The optimal interference form for each distance state has been obtained.Compared with the single-interference decision method,the proposed decision method can effectively improve the success rate of interference.

Attribute Reduction of Hybrid Decision Information Systems Based on Fuzzy Conditional Information Entropy

The presence of numerous uncertainties in hybrid decision information systems(HDISs)renders attribute reduction a formidable task.Currently available attribute reduction algorithms,including those based on Pawlak attribute importance,Skowron discernibility matrix,and information entropy,struggle to effectively manages multiple uncertainties simultaneously in HDISs like the precise measurement of disparities between nominal attribute values,and attributes with fuzzy boundaries and abnormal values.In order to address the aforementioned issues,this paper delves into the study of attribute reduction withinHDISs.First of all,a novel metric based on the decision attribute is introduced to solve the problem of accurately measuring the differences between nominal attribute values.The newly introduced distance metric has been christened the supervised distance that can effectively quantify the differences between the nominal attribute values.Then,based on the newly developed metric,a novel fuzzy relationship is defined from the perspective of“feedback on parity of attribute values to attribute sets”.This new fuzzy relationship serves as a valuable tool in addressing the challenges posed by abnormal attribute values.Furthermore,leveraging the newly introduced fuzzy relationship,the fuzzy conditional information entropy is defined as a solution to the challenges posed by fuzzy attributes.It effectively quantifies the uncertainty associated with fuzzy attribute values,thereby providing a robust framework for handling fuzzy information in hybrid information systems.Finally,an algorithm for attribute reduction utilizing the fuzzy conditional information entropy is presented.The experimental results on 12 datasets show that the average reduction rate of our algorithm reaches 84.04%,and the classification accuracy is improved by 3.91%compared to the original dataset,and by an average of 11.25%compared to the other 9 state-of-the-art reduction algorithms.The comprehensive analysis of these research results clearly indicates that our algorithm is highly effective in managing the intricate uncertainties inherent in hybrid data.

Service Function Chain Deployment Algorithm Based on Multi-Agent Deep Reinforcement Learning

Aiming at the rapid growth of network services,which leads to the problems of long service request processing time and high deployment cost in the deployment of network function virtualization service function chain(SFC)under 5G networks,this paper proposes a multi-agent deep deterministic policy gradient optimization algorithm for SFC deployment(MADDPG-SD).Initially,an optimization model is devised to enhance the request acceptance rate,minimizing the latency and deploying the cost SFC is constructed for the network resource-constrained case.Subsequently,we model the dynamic problem as a Markov decision process(MDP),facilitating adaptation to the evolving states of network resources.Finally,by allocating SFCs to different agents and adopting a collaborative deployment strategy,each agent aims to maximize the request acceptance rate or minimize latency and costs.These agents learn strategies from historical data of virtual network functions in SFCs to guide server node selection,and achieve approximately optimal SFC deployment strategies through a cooperative framework of centralized training and distributed execution.Experimental simulation results indicate that the proposed method,while simultaneously meeting performance requirements and resource capacity constraints,has effectively increased the acceptance rate of requests compared to the comparative algorithms,reducing the end-to-end latency by 4.942%and the deployment cost by 8.045%.

Discovering Latent Variables for the Tasks With Confounders in Multi-Agent Reinforcement Learning

Efficient exploration in complex coordination tasks has been considered a challenging problem in multi-agent reinforcement learning(MARL). It is significantly more difficult for those tasks with latent variables that agents cannot directly observe. However, most of the existing latent variable discovery methods lack a clear representation of latent variables and an effective evaluation of the influence of latent variables on the agent. In this paper, we propose a new MARL algorithm based on the soft actor-critic method for complex continuous control tasks with confounders. It is called the multi-agent soft actor-critic with latent variable(MASAC-LV) algorithm, which uses variational inference theory to infer the compact latent variables representation space from a large amount of offline experience.Besides, we derive the counterfactual policy whose input has no latent variables and quantify the difference between the actual policy and the counterfactual policy via a distance function. This quantified difference is considered an intrinsic motivation that gives additional rewards based on how much the latent variable affects each agent. The proposed algorithm is evaluated on two collaboration tasks with confounders, and the experimental results demonstrate the effectiveness of MASAC-LV compared to other baseline algorithms.

Attribute Reduction Method Based on Sequential Three-Branch Decision Model

Attribute reduction is a research hotspot in rough set theory. Traditional heuristic attribute reduction methods add the most important attribute to the decision attribute set each time, resulting in multiple redundant attribute calculations, high time consumption, and low reduction efficiency. In this paper, based on the idea of sequential three-branch decision classification domain, attributes are treated as objects of three-branch division, and attributes are divided into core attributes, relatively necessary attributes, and unnecessary attributes using attribute importance and thresholds. Core attributes are added to the decision attribute set, unnecessary attributes are rejected from being added, and relatively necessary attributes are repeatedly divided until the reduction result is obtained. Experiments were conducted on 8 groups of UCI datasets, and the results show that, compared to traditional reduction methods, the method proposed in this paper can effectively reduce time consumption while ensuring classification performance.

Method for triangular fuzzy multiple attribute decision making based on two-dimensional density operator method

Aiming at the triangular fuzzy(TF)multi-attribute decision making(MADM)problem with a preference for the distribution density of attribute(DDA),a decision making method with TF number two-dimensional density(TFTD)operator is proposed based on the density operator theory for the decision maker(DM).Firstly,a simple TF vector clustering method is proposed,which considers the feature of TF number and the geometric distance of vectors.Secondly,the least deviation sum of squares method is used in the program model to obtain the density weight vector.Then,two TFTD operators are defined,and the MADM method based on the TFTD operator is proposed.Finally,a numerical example is given to illustrate the superiority of this method,which can not only solve the TF MADM problem with a preference for the DDA but also help the DM make an overall comparison.

A survey on multi-agent reinforcement learning and its application

Multi-agent reinforcement learning(MARL)has been a rapidly evolving field.This paper presents a comprehensive survey of MARL and its applications.We trace the historical evolution of MARL,highlight its progress,and discuss related survey works.Then,we review the existing works addressing inherent challenges and those focusing on diverse applications.Some representative stochastic games,MARL means,spatial forms of MARL,and task classification are revisited.We then conduct an in-depth exploration of a variety of challenges encountered in MARL applications.We also address critical operational aspects,such as hyperparameter tuning and computational complexity,which are pivotal in practical implementations of MARL.Afterward,we make a thorough overview of the applications of MARL to intelligent machines and devices,chemical engineering,biotechnology,healthcare,and societal issues,which highlights the extensive potential and relevance of MARL within both current and future technological contexts.Our survey also encompasses a detailed examination of benchmark environments used in MARL research,which are instrumental in evaluating MARL algorithms and demonstrate the adaptability of MARL to diverse application scenarios.In the end,we give our prospect for MARL and discuss their related techniques and potential future applications.