Information Technology Reference
In-Depth Information
be used with both languages and made it possible to rapidly develop a number of pro-
totype agents with those languages. Through these prototypes, we were able to identify
that AF-AgentSpeak was suited to the Leader role, this is due to the comprehensive
planning support available, and AF-TeleoReactive was suited to the Herder role, as it is
designed to react quickly to a changing environment.
This pilot evaluation serves to justify the comparability of the created languages. As
this project was completed by members of the team responsible for the creation of the
framework, evaluation of the CLFs ease of use is currently future work. This thorough
evaluation will be performed when the Agent-Oriented Software Engineering course
runs again (Jan 2012). The planned evaluation similarly allows the assessment of the
CLF in terms only of the usability and comparability of the produced languages, whilst
this evaluation is useful it is not sufficient to validate the goal of providing a common
toolset that can be adapted to different agent models.
To properly validate the usefulness of the components for the development of AOP
languages would require a large number of people creating languages both with and
without the support of the framework. Whilst ideally we would like the feedback this
would bring, for logistical reasons alone we will not be attempting it.
5
Conclusions
Our main objective making these changes to Agent Factory is to develop versions of
existing AOP languages for Agent Factory, that are adapted to employ a consistent un-
derlying infrastructure which we hope will allow developers to focus on understanding
the strengths and weaknesses of the languages rather than the supporting machinery. By
providing this common infrastructure and implementing various AOP languages, we
also hope to gain additional insight into the weaknesses of the current state-of-the-art
in AOP with the goal of identifying and exploring potential features that will underpin
the next generation agent programming languages. Finally, we hope that Agent Factory
will help to foster the development of new AOP languages by reducing the complexity
of AOP language development.
Acknowledgements.
This work was supported, in part, by Science Foundation Ireland
grant 07/CE/I1147 to CLARITY: Centre for Sensor Web Technologies and by Envi-
ronmental Protection Agency Ireland project number 2008-WRM-MS-1-S1, WAIST -
Waste Augmentation and Integrated Shipment Tracking, funded under STRIVE- Waste,
Resource Management and Chemicals Call 2007.
References
1. Bellifemine, F.L., Caire, G., Greenwood, D.: Developing multi-agent systems with JADE.
Wiley, Chichester (2007)
2. Bordini, R.H., Hubner, J.F., Wooldridge, M.J.: Programming multi-agent systems in AgentS-
peak using Jason. Wiley Interscience, Hoboken (2007)
3. Collier, R.W., O'Hare, G.M.P.: Modeling and programming with commitment rules in agent
factory. In: Giurca, Gasevic, Tavater (eds.) Handbook of Research on Emerging Rule-Based
Languages and Technologies: Open Solutions and Approaches. IGI Publishing (2009)
