Making Decision with Reactive Multi-agent Systems: A Possible Alternative to Regular Decision Processes for Platoon Control Issue

Abstract

Decision processing is a key element in computer science, in automatic control and in robotics. The literature presents a lot of various approaches for decision processing. These approaches generally depend on both the adopted conceptual point of view and of the application field. Some of the classical methods suffers from several problems such as a limited adaptivity, or a high computational cost. In this context, reactive multi-agent systems can be good candidates to overcome some of these drawbacks. The goal of this paper is to present how to use reactive multi-agent systems to make decision, what are their advantages and their drawbacks as compared to classical methods. It also illustrates the proposal with an example of application dealing with the platoon control issue.