Evaluation of Multi-Agent Platforms for Ubiquitous Computing

Evaluation of Multi-Agent Platforms for Ubiquitous Computing

ABSTRACT

Ubiquitous Computing can be described as the third stage in the computing history where every user is surrounded by many “computers”. This paper provides an evaluation of a number of multi-agent platforms to decide their appropriateness as an infrastructure for ubiquitous computing. To decide upon what criteria to use as evaluation criteria , a small literature study on the field of ubiquitous computing is performed. The identified criteria are then used to rank the different platforms. To verify parts of the result one criterion is further investigated through an experiment.

Keywords: Ubiquitous Computing, Multi-Agent platforms, Ambient Intelligence.

INTRODUCTION

This chapter begins with a brief introduction of the central topics in Ubiquitous Computing and multi-agent platforms. The presentation of the goals follows after that, and then the chapter ends with an explanation of the structure of this study. In the last couple of months it has become more and more frequent to hear and read about everyday objects that connect to each other. For example MP3 players that includes an inbuilt FM radio-transmitter, making it possible to listen to your MP3 music-files through your car-radio. Other examples are the connection of a PC to your TV-set, making it possible to record the program when you are leaving for a visit to the bathroom and continue at the same spot in the program when you come back. This indicates that the concept of everywhere appearing computers is not any longer a phenomenon in the future but in fact something that is here today. The term Ubiquitous means “everywhere appearing” and Ubiquitous Computing can bee seen as the next generation of computing where every user is surrounded by a multitude of computers embedded in everyday objects. These computers will be connected to each other to provide user applications and information everywhere. In order to use them, software agents are frequently mentioned due to their characteristics such as mobility and autonomy (tries to achieve their own goals without user interaction). Multi-agent platforms are used to develop and run such agents. However, when looking at standard multi-agent platforms it is obvious that many of them are constructed to be used on ordinary computers, connected to each other with fixed wired networks. There are of course other techniques like web-services and plug and play that could be used in ubiquitous computing but in this thesis we look at multiagent platforms.

1.2 Goals and aims

The aim of this thesis is to investigate which multi-agent platforms that are suitable for use in Ubiquitous Computing. A secondary goal is to give suggestions about general characteristic demands on platforms to be useful in Ubiquitous Computing. In order to reach our aims we first try to answer the following questions:

i.        What criteria are there for multi-agent platforms in Ubiquitous Computing?

ii.        What multi-agent platforms are there?

 iii.        What platforms meet the criteria?

To answer the first question we perform a literature study on promising areas for Ubiquitous Computing to identify criteria that multi-agent platforms must meet to be appropriate for Ubiquitous Computing applications. We choose three different areas of Ubiquitous Computing and gather together a set of criteria that we consider characteristic for the field of Ubiquitous Computing. The next step provides a list of multi-agent platforms to bee evaluated. The multi-agent platforms are then evaluated to what degree they fulfil the criteria found in the literature study. After that the most promising platforms are experimentally evaluated running a small application with a selection of the Ubiquitous Computing characteristics.

Evaluation of Multi-Agent Platforms for Ubiquitous Computing

Chapters: 1 - 5
Delivery: Email

Number of Pages: 75

Price: 5000 NGN

In Stock

Buy the Complete Material Now