Multi-agent System Architecture Research Articles

Decentralized, norm-free, and adaptive event-triggered distributed control of nonholonomic mobile robots

The overarching contribution of this paper is a novel event-triggered distributed control architecture for multiagent systems that are composed of a team of nonholonomic mobile robots. Specifically, the robot dynamics are first feedback linearised to equivalently represent each robot with double integrator dynamics and the event-triggering law is then presented. This law is characterised by being decentralised (depending only on own signals of a robot), norm-free (independent of distances for reducing robot-to-robot data transmissions), and adaptive (estimating signals without using global information). Moreover, its system-theo- retical analysis is provided, which holds for both the sampled data exchange method and the solution-predictor curve-based data exchange method. In contrast to the well-studied sampled data exchange, the solution-predictor curve method further reduces robot-to-robot data transmissions by making each agent to store a generic form of this curve and exchange its parameters when an event occurs for approximating the solution trajectory of each robot. Finally, an illustrative numerical example is included to demonstrate the overall architecture.

Multi-agent system architecture for winter road maintenance: a real Spanish case study

Road safety remains a critical issue in contemporary society, where the sudden deterioration of road conditions due to weather-related natural phenomena poses significant risks. These abrupt changes can lead to severe safety hazards on the roads, making real-time monitoring and control essential for maintaining road safety. In this context, technological advancements, especially in sensor networks and intelligent systems, play a fundamental role in efficiently managing these challenges. This study introduces an innovative approach that leverages a sophisticated sensor platform coupled with a multi-agent system. This integration facilitates the collection, processing, and analysis of data to preemptively determine the appropriate chemical treatments for roads during severe winter conditions. By employing advanced data analysis and machine learning techniques within a multi-agent framework, the system can predict and respond to adverse weather effects swiftly and with a high degree of accuracy. The proposed system has undergone rigorous testing in a real-world environment, which has verified its operational effectiveness. The results from the deployment of the multi-agent architecture and its predictive capabilities are encouraging, suggesting that this approach could significantly enhance road safety in extreme weather conditions. Furthermore, the proposed architecture allows the system to evolve and scale over time. This paper details the design and implementation of the system, discusses the results of its field tests, and explores potential improvements.

Dynamic scheduling mechanism for intelligent workshop with deep reinforcement learning method based on multi-agent system architecture

With the development and changes of industry and market demand, the personalized customization production mode with small batch and multiple batches has gradually become a new production mode. This makes production environment become more complex and dynamic. However, traditional production workshops cannot effectively adapt to this environment. Combining with new technologies, transforming traditional workshops into intelligent workshop to cope with new production mode become an urgent problem. Therefore, this paper proposes a multi-agent manufacturing system based on IoT for intelligent workshop. Meanwhile, this paper takes flexible job shop scheduling problem (FJSP) as a specific production scenario and establishes relevant mathematics model. To build the agent in intelligent workshop, this paper proposes a data-based with combination of virtual and physical agent (DB-VPA) which has information layer, software layer and physical layer. Then, based on the manufacturing system, this paper designs a dynamic scheduling mechanism with deep reinforcement learning (DRL) for intelligent workshop. This method contains three aspects: (1) Modeling production process based on Markov decision process (MDP). (2) Designing communication mechanism for DB-VPAs. (3) Designing scheduling model combining with improved genetic programming and proximal policy optimization (IGP-PPO) which is a DRL method. Finally, relevant experiments are executed in a prototype experiment platform. The experiments indicate that the proposed method has superiority and generality in solving scheduling problem with dynamic events.

Food Passports and Intelligent Food Recipes: The Data-Oriented Way of Producing Food

This paper discusses a Multiagent System architecture and its supporting concepts and technical implementation for food-producing systems. The Food and Beverages industry is in dire need of embracing more advanced industrial digitization practices to be able to meet immediate sustainability requirements. One such challenge is improving and optimizing resource utilization. A more rationalized use of raw materials, production processes and machinery is a key aspect of the latter. At the same time, food is still produced, in many instances, in a very traditional way, particularly when it comes to ingredient selection. The architecture introduced in this paper has the ambition of articulating the supply and demand of food products based in the technical characteristics of food (e.g., composition, nutritional content and value, processing efficiency, and taste profiles) instead of the more traditional characteristics. To that end, the paper introduces, among other key concepts, the notion of the Digital Food Identity Card as an active digital document that stores such technical information and the notion of Dynamic Recipe as an active software entity with the dynamic capability of adjusting production parameters in a recipe to market conditions, as well as coordinating the resources included in the execution of a industrial food recipe. Both concepts are modeled as agents and part of the proposed architecture. Finally, the paper provides a formal grammar for a domain specific language, that can be used by human specialists to define a dynamic recipe and be simultaneously interpreted by Dynamic Recipe Agents to enact the behaviours previously described. The domain specific language is an integral part of the architecture, together with the agent-based formulation, and an important link in creating an effectively data-oriented way of articulating different stake-holder concerns. The proposed solution is informed by requirements elicited within domain experts and sustainability targets that food-producing companies currently must adhere to.

Towards a Docker-based architecture for open multi-agent systems

<p>In open multi-agent systems (OMAS), heterogeneous agents in different environments or models can migrate from one system to another, taking their attributes and knowledge and increasing developing complexity compared to conventional multi-agent systems (MAS). Furthermore, the complexity of opening may be due to the uncertainties and dynamic behavior that the change of agents entails, needing to formulate techniques to analyze this complexity and understand the system’s global behavior. We used Docker to approach these problems and make the architecture flexible to handle distinct types of programming languages and frameworks of agents. This paper presents a Docker-based architecture to aid OMAS development, acting on agent migration between different models running in heterogeneous hardware and software scenarios. We present a simulation scenario with NetLogo’s Open Sugarscape 2 Constant Growback and JaCaMo’s Gold Miners to verify the proposal’s feasibility.</p>

Formal model of multi-agent architecture of a software system based on knowledge interpretation

The use of agents across diverse domains within computer science and artificial intelligence is experiencing a notable surge in response to the imperatives of adaptability, efficiency, and scalability. The subject of this study is the application of formal methods to furnish a framework for knowledge interpretation with a specific focus on the agent-based paradigm in software engineering. This study aims to advance a formal approach to knowledge interpretation by leveraging the agent-based paradigm. The objectives are as follows: 1) to examine the current state of the agent-based paradigm in software engineering; 2) to describe the basic concepts of the knowledge interpretation approach; 3) to study the general structure of the rule extraction task; 4) to develop the reference structure of knowledge interpretation; 5) to develop a multi-agent system architecture; 6) and to discuss the research results. This study employs formal methods, including the use of closed path rules and predicate logic. Specifically, the integration of closed path rules contributes to the extraction and explication of facts from extensive knowledge bases. The obtained results encompass the following: 1) a rule mining approach grounded in closed path rules and tailored for processing extensive datasets; 2) a formalization of relevance that facilitates the scrutiny and automated exclusion of irrelevant fragments from the explanatory framework; and 3) the realization of a multi-agent system predicated on the synergy among five distinct types of agents, dedicated to rule extraction and the interpretation of acquired knowledge. This paper provides an example of the application of the proposed formal tenets, demonstrating their practical context. The conclusion underscores that the agent-based paradigm, with its emphasis on decentralized and autonomous entities, presents an innovative framework for handling the intricacies of knowledge processing. It extends to the retrieval of facts and rules. By distributing functions across multiple agents, the framework offers a dynamic and scalable solution to effectively interpret vast knowledge repositories. This approach is particularly valuable in scenarios where traditional methods may struggle to cope with the volume and complexity of information.

Hessian Distributed Ant Optimized Perron–Frobenius Eigen Centrality for Social Networks

Terabytes of data are now being handled by an increasing number of apps, and rapid user decision-making is hampered by data analysis. At the same time, there is a rise in interest in big data analysis for social networks at the moment. Thus, adopting distributed multi-agent-based technology in an optimum way is one of the solutions to effective big data analysis for social networks. Studying the development of a social network helps users gain an understanding of interactions and relationships and guides them in making decisions. In this study, a method called Hessian Distributed Ant Optimized and Perron–Frobenius Eigen Centrality (HDAO-PFEC) is developed to analyze large amounts of data (i.e., Big Data) in a computationally accurate and efficient manner. Designing an adaptable Multi-Agent System architecture for large data analysis is the primary goal of HDAO-PFEC. Initially, using a Hessian Mutual Distributed Ant Optimization MapReduce model, comparable user interest tweets are produced in a computationally efficient manner. Eigen Vector Centrality is a measure of a node’s importance in a network (i.e., a social network), which allows association with other significant nodes (i.e., users), allowing for a greater effect on social networks. With this goal in mind, a MapReduce methodology in the Hadoop platform using Big Data, which enables quick and ordered calculations, is used in a distributed computing method to estimate the Eigen Vector Centrality value for each social network member. Lastly, extensive investigative experimental learning demonstrates the HDAO-PFEC method’s use and accuracy as well as its time and overhead on the well-known sentiment 140 dataset.

ManuChain II: Blockchained Smart Contract System as the Digital Twin of Decentralized Autonomous Manufacturing Toward Resilience in Industry 5.0

In Industry 5.0 vision, machines are empowered with the interaction capability to autonomously make local decisions and coordinate with each other as well as humans. However, how to form a group consensus on the rapid self-organizing of the manufacturing process is critical for achieving manufacturing resilience under disturbances and disruptions. Based on our formerly developed system ManuChain (Leng et al., 2020), this article proposes a blockchained smart contract system (BSCS), named ManuChain II, as the digital twin of a decentralized autonomous manufacturing system for achieving resilience in Industry 5.0. A blockchain-secured multiagent system architecture together with a product data model is established to form the BSCS. The BSCS could prevent tampering with data and enhance the transparency of the product manufacturing process. In BSCS, two types of blockchained smart contracts (SCs) are established with a bi-level interplay computing architecture. The lower-level contracts perform the predefined and learned patterns of task coordination for achieving resilience under internal disruptions. The upper-level contracts are incorporated with communication-efficient decentralized deep-learning algorithms for the learning, updating, transferring, and sharing of coordination patterns used in the autonomous decision in lower-level SCs, thereby achieving the continuous improvement of the system’s decentralized autonomous intelligence. Via incorporating the decentralized deep learning algorithms into blockchained SCs, this study reveals a new way to realize the self-organizing intelligence of the manufacturing system for enhancing resilience toward Industry 5.0.

A novel actor–critic–identifier architecture for nonlinear multiagent systems with gradient descent method

The infinite-horizon optimal consensus control problem for continuous-time unknown nonlinear multiagent systems (MASs) is solved via an online adaptive reinforcement learning approach. Three different neural network (NN) structures are proposed as part of a novel actor–critic–identifier (ACI) to approximate the Hamilton–Jacobi–Bellman (HJB) equations. The actor and critic NNs approximate the optimal controls and optimal value functions, respectively, and identifier NN approximates the unknown system model. The ACI architecture is simpler than most known architectures, which has the benefit of allowing actor, critic, and identifier learning to occur concurrently and continuously without the system model. It is important to note that the tuning laws for the three different NNs are designed using gradient descent method. Adaptive control techniques based on Lyapunov method are used to examine the algorithm’s convergence. To verify the viability of the proposed approach, we offer a simulated example.

Reactive synthesis for relay-explorer consensus with intermittent communication

A distributed multi-agent system architecture is explored to reach approximate consensus with intermittent communication. The multi-agent system is cast as a relay-explorer problem, where a relay agent intermittently provides navigational feedback to multiple explorer agents that do not have on-board absolute navigational sensors in a pre-defined sub-region. Within each sub-region, there is one relay agent responsible for servicing the corresponding explorer agents, and the estimated trajectory of an explorer agent can cross the boundary and enter another sub-region. We develop a reactive synthesis approach to formulate the mission specifications, while the state–space system dynamics provide real-time information for state corrections. Specifically, we pre-synthesize a set of planning strategies corresponding to candidate instantiations (i.e., pre-specified representative information scenarios) to dynamically switch among the explorers, and the planning strategies enable transfer of the servicing responsibility between relay agents. To guarantee stability of the switching strategies and the approximate consensus of the explorer agents, we develop maximum dwell-time conditions using a Lyapunov-based analysis to allow the explorer agents to drift for a pre-defined period without requiring servicing from the relay agents. Finally, we include a simulation study to demonstrate the performance of the developed method.

Engineering IoT-Based Open MAS for Large-Scale V2G/G2V

In this paper, we aimed to demonstrate how to engineer Internet of Things (IoT)-based open multiagent systems (MASs). Specifically, we put forward an IoT/MAS architectural framework, along with a case study within the important and challenging-to-engineer vehicle-to-grid (V2G) and grid-to-vehicle (G2V) energy transfer problem domain. The proposed solution addresses the important non-functional requirement of scalability. To this end, we employed an open multiagent systems architecture, arranging agents as modular microservices that were interconnected via a multi-protocol Internet of Things platform. Our approach allows agents to view, offer, interconnect, and re-use their various strategies, mechanisms, or other algorithms as modular smart grid services, thus enabling their seamless integration into our MAS architecture, and enabling the solution of the challenging V2G/G2V problem. At the same time, our IoT-based implementation offers both direct applicability in real-world settings and advanced analytics capabilities via enabling digital twin models for smart grid ecosystems. We have described our MAS/IoT-based architecture in detail; validated its applicability via simulation experiments involving large numbers of heterogeneous agents, operating and interacting towards effective V2G/G2V; and studied the performance of various electric vehicle charging scheduling and V2G/G2V-incentivising electricity pricing algorithms. To engineer our solution, we used ASEME, a state-of-the-art methodology for multiagent systems using the Internet of Things. Our solution can be employed for the implementation of real-world prototypes to deliver large-scale V2G/G2V services, as well as for the testing of various schemes in simulation mode.

A distributed simplex architecture for multi-agent systems

We present the Distributed Simplex Architecture (DSA), a new runtime assurance technique that provides safety guarantees for multi-agent systems (MASs). DSA is inspired by the Simplex control architecture of Sha et al., but with some significant differences. The traditional Simplex approach is limited to single-agent systems or a MAS with a centralized control scheme. DSA addresses this limitation by extending the scope of Simplex to include MASs under distributed control. In DSA, each agent runs a local instance of traditional Simplex such that the preservation of safety in the local instances implies safety for the entire MAS. Control Barrier Functions (CBFs) play a critical role. They are used to define DSA’s core components – the baseline controller and the decision module’s logic for switching between advanced and baseline control – and they provide the basis for the proof of safety. We present a general proof of safety for DSA, provided the CBF-related optimization problem solved by the baseline controller is feasible (has a solution) at each time step for which the baseline controller is in control. We also propose a novel extension to the switching logic designed to avoid states in which this optimization problem is infeasible. Finally, we present experimental results for several case studies, including flocking with collision avoidance, safe navigation of ground rovers through way-points, and safe operation of a microgrid.

A multi-agent system based architecture for enabling Edge autonomous management

Edge computing extends the Cloud computing paradigm by providing computation resources at the edge of the network where data is being generated to effectively meet the demands of real-time or latency-sensitive applications. Given Edge’s particular characteristics, centralized controlling and decision making in the Cloud are no longer the only option. In contrast, it should be enabled throughout the Edge deployment’s hierarchy from the top level of the hierarchy to near the IOT devices (at the edge of the network). This means enabling and enhancing intelligence and autonomy at the edge. This paper suggests an autonomous vision for Edge management. We propose a multi-agent system architecture, enabling autonomous decision making at Edge environments. A case study, using learning agents, is presented to illustrate the way the proposed solution enables sound management decisions. The case study focuses on smart offloading and autonomous power management. The profitability of enhancing intelligence and autonomy is proved in both cases.

An AI-Layered with Multi-Agent Systems Architecture for Prognostics Health Management of Smart Transformers: A Novel Approach for Smart Grid-Ready Energy Management Systems

After the massive integration of distributed energy resources, energy storage systems and the charging stations of electric vehicles, it has become very difficult to implement an efficient grid energy management system regarding the unmanageable behavior of the power flow within the grid, which can cause many critical problems in different grid stages, typically in the substations, such as failures, blackouts, and power transformer explosions. However, the current digital transition toward Energy 4.0 in Smart Grids allows the integration of smart solutions to substations by integrating smart sensors and implementing new control and monitoring techniques. This paper is proposing a hybrid artificial intelligence multilayer for power transformers, integrating different diagnostic algorithms, Health Index, and life-loss estimation approaches. After gathering different datasets, this paper presents an exhaustive algorithm comparative study to select the best fit models. This developed architecture for prognostic (PHM) health management is a hybrid interaction between evolutionary support vector machine, random forest, k-nearest neighbor, and linear regression-based models connected to an online monitoring system of the power transformer; these interactions are calculating the important key performance indicators which are related to alarms and a smart energy management system that gives decisions on the load management, the power factor control, and the maintenance schedule planning.

Blockchained smart contract pyramid-driven multi-agent autonomous process control for resilient individualised manufacturing towards Industry 5.0

The production control for the mass individualisation paradigm of R&D-stage products is challenging due to the mix-flow and frequently-disturbed environment. With the convergence of the sustainable development goals and the increasing individualised demands in products, resilient manufacturing is envisioned in Industry 5.0 proposition. Concerning that conventional centralised production control methods suffer from low stability and inefficiency of decisions under frequent disruptions, this paper establishes a blockchained smart contract pyramid-driven multi-agent autonomous process control (BSCP-MAAPC) approach for improving the timeliness and adaptability of control towards resilient individualised manufacturing. Firstly, a blockchain-based multi-agent system architecture is designed based on agent encapsulation of manufacturing units. Blockchained smart contracts are used as the enabler of the multi-agent system for peer-to-peer negotiation and coordination of tasks. Secondly, a quad-play blockchained smart contract pyramid together with a series of decentralised control patterns are designed to enable the initial task dispatching of various individualised demands, as well as rapid dynamic adjustment of schedule in response to internal random disruptions. Finally, a blockchained smart contract pyramid-driven multi-agent autonomous process control system prototype is built in the ManuChain system, and experiments are conducted to analyze the proposed BSCP-MAAPC approach in different environments. HIGHLIGHTS A blockchained smart contract pyramid-driven multi-agent autonomous process control (BSCP-MAAPC) approach. A quad-play blockchained smart contract pyramid together with a series of decentralized control patterns. Blockchained smart contracts as the enabler of the multi-agent system for peer-to-peer negotiation and coordination of tasks.

A CPPS-architecture and workflow for bringing agent-based technologies as a form of artificial intelligence into practice

Abstract Due to the increase in Artificial Intelligence in the production systems domain, Industry 4.0 (I4.0) experts must collaborate with autonomous systems. Industrial AI raises several concerns about existing standards, which provide guidelines and design patterns. One way to realize I4.0 systems are Industrial Agents (IAs) due to their inherent autonomy and collaboration. Multi-Agent Systems (MASs) are well suited for realizing distributed AI in I4.0 components. Considering the properties of IAs and existing standards, an MAS architecture is presented for flexible and intelligent Cyber-Physical Production Systems. The article compares I4.0 standardization efforts relevant to adapt AI in the form of IAs, illustrates how IA design patterns can be used, and introduces the Multi-Agent aRchitecture for Industrial Automation applying desigN patterNs practicEs “MARIANNE”. An implementation guideline is presented to put this CPPS into practice.

Архитектура многоагентных систем для федеративного обучения

The concept of federated learning has become widespread in working with data, mainly due to the fact that it allows training on data directly on the nodes where they are stored. As a result, no data transfer is required. After the training is completed on each node, only the trained model is transmitted to the central server for aggregation. Multi-agent systems behave in a similar way, because agents allow you to train machine learning models on local devices, while preserving confidential information. The ability of agents to interact with each other makes it possible to generalize (aggregate) such models and reuse them. This article presents the architecture of multi-agent systems for federated learning. It highlights the elements that make up the agent platform and the structure of the JADE platform. Describes the lifecycle of all agents used to perform a full training cycle in the MAC\_FL environment. The configurations of agent placement for each of the proposed architectures of multi-agent systems of federated learning are analyzed and described: centralized, decentralized and hierarchical.

Converging IoT protocols for the data integration of automation systems in the electrical industry

The Internet of Things (IoT) plays an important role in the development of applications for the Electrical Industry. The data has become essential for the technological advances in this industry, changing abruptly due to the distributed energy resources and the smart grid concept. The data integration also represents a relevant matter to address when performing IoT projects for electrical industry applications. The electrical industry requires incorporating different IoT protocols to comply with the distinct scenarios and the interoperability. This work proposes integrating the HTTP REST, MQTT, LoRaWAN, and OPC UA open communications protocols into an IoT platform and interoperable architecture for smart grid applications. This approach intends to contribute to smart grid solutions and automation systems in the electrical industry, promotes further integration of communication protocols for this field, and enables additional data-based applications for their automation systems. A Multi-Agent System architecture based on the IoT integration is proposed as an application to show how implementable is this approach in current and technological electrical systems.

Towards a formal model of computation for RMAS

RMAS is a multi-agent system architecture and associated model of computation that has been recently proposed as promising framework for autonomic computation and as viable implementation means for industrial agents. RMAS is proposed as an effective tool in the realization of the full-fledged concept of cyber-physical systems of systems, in agreement with new visions of the digital future and prominent standardization efforts like the IEC 61499. Currently, there is the need to establish a solid formalism for the model of computation of RMAS in order to harness and fully express its possibilities. In particular, this model should bridge the disciplines of control and systems engineering and computer science, whilst concerning artificial intelligence. This work proposes a first operational step in this direction. A formalism of the RMAS model of computation is established for the first time, and a complete definition of hierarchical nested levels of RMAS units is proposed. This work constitutes the ground for industry-ready technological instances of the RMAS architecture.

Optimization of the Structure of Power System Multi-Agent Control

The article provides an overview of the application of multi-agent control, provides a classification of multi-agent system architectures, and analyzes existing methods for constructing the structure of multi-agent control in the electric power industry. A method for constructing an optimal structure of multi-agent optimal control is proposed. A field of applications of proposed method for a multi-agent voltage and reactive power optimal control in a power system is discussed. The proposed method of optimal design of multi-agent power system control structure can be applied to various objective function of optimization. The application of multi-agent control for automatic voltage and reactive power control in the power system is currently being researched.