China
Africa
Explore chapters and articles related to this topic
Conceptual Multi-Agent System Design for Distributed Scheduling Systems
Published in Vijaya Kumar Manupati, Goran D. Putnik, Maria Leonilde Rocha Varela, Smart and Sustainable Manufacturing Systems for Industry 4.0, 2023
Filipe Alves, Ana Maria A. C. Rocha, Ana I. Pereira, Paulo Leitão
The agent concept does not have a consensus or unique definition, mainly due to its features and attributes. Some proposed definitions that can be found in the literature are: “An agent is a persistent computation that can perceive its environment and reason and act both alone and with other agents. The key concepts in this definition are interoperability and autonomy” (Singh, 1998).“An agent is a computational entity that can be viewed as perceiving and acting upon its environment, that is autonomous and that operates flexibly and rationally in a variety of environmental circumstance” (Weiss, 1999).“An agent is a computer system that is situated in an environment and that is capable of autonomous action in this environment in order to meet its design objectives” (Wooldridge, 2009).
Cooperative Robotic Systems in Agriculture
Published in Dan Zhang, Bin Wei, Robotics and Mechatronics for Agriculture, 2017
Granted that an advanced RA has better performance and is more robust than a simple RA, however, researchers have ratiocinated that a cooperative team of simple RAs has better accuracy in localization, navigation, path planning, and optimal performance. The combination of two or more interacting intelligent agents is referred to as a multi-agent system. A multi-agent system is a swarm intelligence system having a smart team of agents effectively interacting with each other to complete common tasks. Multi-agent systems have the ability to resolve complicated tasks which are difficult or impossible for a single agent to accomplish (Barca and Sekercioglu, 2013). Multi-agent systems in cooperative environments allow more sophisticated agents to share their capabilities with other agents which have limited capabilities (Bailey et al., 2011). Multi-agent systems have the robustness of a single agent; thus, they have been applied successfully to agriculture and manufacturing applications.
DAI for Document Retrieval
Published in Satya Prakash Yadav, Dharmendra Prasad Mahato, Nguyen Thi Dieu Linh, Distributed Artificial Intelligence, 2020
Anuj Kumar, Satya Prakash Yadav, Sugandha Mittal
The key idea utilized in DPS and MABS is the program called machine operators for reflection. A specialist is a self-sufficient virtual (or physical) object that comprehends its condition and follows up on it. In a similar procedure, a specialist is generally ready to team up with different operators to accomplish a typical target that one operator couldn’t accomplish alone. This strategy for correspondence utilizes a language for speaking with the beneficiary. A valuable first grouping is to isolate the specialists into: Reactive agent: not much more than an automaton that receives input, processes it, and generates output.Deliberative agent: by contrast, the deliberative agent should have an unbiased view of his environment and be in a position to execute his plans.Reactive agent: not much more than a car that receives input, processes it, and generates output.Deliberative agent: a deliberative agent will, on the contrary, have an internal perception of its world and be able to implement its plans.Hybrid agent: a hybrid agent is a mixture of reactive and deliberative, implementing their plans but also reacting directly without deliberation to external events.
Water distribution in community irrigation using a multi-agent system
Published in Journal of the Royal Society of New Zealand, 2023
Kitti Chiewchan, Patricia Anthony, Birendra KC, Sandhya Samarasinghe
An ‘agent’ is a special software component that provides an interoperable interface to an autonomous system in the same way that a human agent works with a client to achieve a particular goal. An agent can work as a solitary agent within a particular environment or interact with other agents if necessary (Wooldridge and Jennings 1995). Agents have been widely used in several technologies such as artificial intelligence (AI), databases and computer networks (Niazi and Hussain 2011). An agent is autonomous, social, proactive and reactive. ‘Autonomous’ means that agents can interoperate without human intervention. An agent has control over its actions and internal state. ‘Social’ means that an agent can work with other agents or humans to achieve their tasks. An agent is ‘reactive’ because it can perceive and automatically respond to changes in its environment. An agent is ‘proactive’ not only because it acts in response to its environment, but it exhibits goal-directed behaviour by taking initiatives (Bellifemine et al. 2007).
Transparent-AI Blueprint: Developing a Conceptual Tool to Support the Design of Transparent AI Agents
Published in International Journal of Human–Computer Interaction, 2022
Zhibin Zhou, Zhuoshu Li, Yuyang Zhang, Lingyun Sun
Artificial intelligence (AI) agents are becoming pervasive and are increasingly incorporated into every aspect of our lives. The term AI agents refers to virtual systems like recommender systems and physical products like intelligent robots or autonomous vehicles (Defense Science Board, 2016) that perceive the environment and take actions to achieve a goal through the use of AI algorithms. Benefiting from increasingly sophisticated machine learning (ML) techniques in recent years, these AI agents have gained autonomy to have a degree of self-governing and self-directed behavior (Anderson et al., 2015; Hancock, 2017; Schaefer et al., 2016). Such agents can perceive the environment and react to it, and perform tasks without human intervention. Additionally, these agents can evolve their ability through learning from given data (e.g., images with labels and human feedback), and have the potential to manage complex situations that could only be solved previously by human intelligence. Because the intelligence of AI agents is not based on pre-set rules, their behavior is not directly predictable. Although intuitive ways of interacting with AI agents have been developed, the underlying algorithms are growing in complexity and therefore decreasing the agents’ understandability. As a result, it is challenging for humans to understand how AI agents function internally and predict their performance. This poses new challenges to the interaction between AI agents and stakeholders including end users and bystanders, thereby causing human-computer interaction (HCI) issues, such as distrust and confusion about the actions of AI agents.
Consensus of hybrid multi-agent systems with heterogeneous dynamics
Published in International Journal of Control, 2020
Qi Zhao, Yuanshi Zheng, Yunru Zhu
Multi-agent system is a system which is composed of many interconnected autonomous agents. These agents can present an orderly co-movement and behaviour through cooperation and self-organisation at the system level. In other words, the agents of multi-agent system not only can achieve their own behavioural goals but also can cooperate with each other to achieve the common goal. Hence, it can solve the complex large-scale problems in the real world. The rapid development of multi-agent system provides modelling and analytical methods to the research on complex systems. On the other hand, it provides a theoretical basis for practical applications. Examples include the formation control of robotic systems, the cooperative control of unmanned aerial vehicles, the attitude alignment of satellite clusters, the target tracking of sensor networks, the congestion control of communication networks and so on Ren and Beard (2008). When we analyse the multi-agent systems, many mathematical methods can be chosen, just like graph theory, matrix theory, the frequency-domain analysis method and the Lyapunov method, etc.