Abstract
An agent is an autonomous computer system situated in an environment to fulfill a design objective. Multi-Agent Systems aim to solve problems in a flexible and robust way by assembling sets of agents interacting in cooperative or competitive ways for the sake of possibly common objectives. Multi-Agent Systems have been applied to several domains ranging from many industrial sectors, e-commerce, health and even entertainment. Agent-Based Modeling, a sort of Multi-Agent Systems, is a technique used to study complex systems in a wide range of domains. A natural or social system can be represented, modeled and explained through a simulation based on agents and interactions. Such a simulation can comprise a variety of agent architectures like reactive and cognitive agents. Despite cognitive agents being highly relevant to simulate social systems due their capability of modelling aspects of human behaviour ranging from individuals to crowds, they still have not been applied extensively. A challenging and socially relevant domain are the Disaster-Rescue simulations that can benefit from using cognitive agents to develop a realistic simulation. In this paper, a Multi-Agent System applied to the Disaster-Rescue domain involving cognitive agents based on the Belief–Desire–Intention architecture is presented. The system aims to bridge the gap in combining Agent-Based Modelling and Multi-Agent Systems approaches by integrating two major platforms in the field of Agent-Based Modeling and Belief-Desire Intention multi-agent systems, namely, NetLogo and Jason.
Highlights
An agent is a purpose-driven entity that interacts with its environment through its perceptions and actions in a persistent, rational and autonomous way [1,2]
In the analysis presented by Hawe et al [14] based on [15], the authors underlines that the emergency management cycle is composed by four phases: 1
The experiments are focused on the interaction between NetLogo and Jason for the purpose of illustrating the feasibility of the platform to scale regarding the number of agents performing rescue actions
Summary
An agent is a purpose-driven entity that interacts with its environment through its perceptions and actions in a persistent, rational and autonomous way [1,2]. Cognitive agents are based on a more complex design that achieves a more robust architecture They can reason about the state of the environment to reach a goal-driven behaviour helpful to model either individuals or crowds. Despite Agent-Based Models being a sort of MAS, both communities have been evolved in independent ways with sparse mutual interaction [12,13] This situation should be amended in order to exploit the knowledge developed by the entire agent-related community for its own sake. ABM can be benefited from the strategic selection of agent’s behaviours that MASs make available In both cases, BDI architecture is suitable to be used because of its flexibility to define agents with a higher level of knowledge representation and deliberative capabilities exhibiting enough reactivity to cope with dynamic changes.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
