A Multi Agent System Modelling an Intelligent Transport System
Vincenzo Di, Alberto Amato, Domenico Soldo, Antonella Giove · InTech eBooks · 2010
In this work the authors have presented a new ITS, developed in the framework of a research project funded by Apulia Region (Italy), focusing the attention on some aspects related to the route planning module. The proposed route planning module is based on a multi-agent cooperative negotiation paradigm. For this reason, particular attention was paid to the linguistic aspects of the problem and an ontological approach was used to define the communication acts among agents. The achieved results reveal how a multi-agent cooperative negotiation paradigm can be useful in solving the problem of multi-objective route optimization. The proposed approach considers individual agent optimization (the process that each agent uses to choose a new route) and agency optimization, performed according to our negotiation protocol, by which the agents achieve a global agreement. The implementation of a multi-agent paradigm gives great flexibility and scalability to the system. As the proposed real case application shows, this system is able to handle time variant constraint. The presence of the mediator introduces a significant level of scalability into the system. Indeed, if there is the need to tackle a new constraint, it will be sufficient to create a new agent dealing with it. The proposed results demonstrate the suitability of the proposed system as support decision system in route planning problems.