近來agent的觀念散佈在CS的領域，但agent-oriented 相關的平台卻不流行，探討原因發現不是因為agent觀念不適用於CS，而是agent-oriented 相關的平台缺少了易用性及動態性，因此我們提出了以Jade平台延伸的Jadexx期望能讓agent-oriented programming達到流行。

在代理人思考的設計上，Jadexx為了使其易用性和動態性會更好。採用了4個設計,分別為1) 使用Java語言，使熟悉Java的人能輕易上手，而現今Java又非常普及。 2)具體實作出desire 和intention儲存結構，當使用者有BDI觀念時，使用時就會非常容易。 3) 新增一種plan叫做meta plan，為Jadexx實作meta-level reasoning的機制。4)為每個desire定義可變動的優先權，使agent在不同情況下，選取的desire也會有所不同。

最後我們實作了Jadexx的一個範例，並且與Jadex、Jason的易用性及動態性做個比較，發現我們在這兩個功能性上，是比較優異的。

Recently, the concept of agent appears in the CS field, but agent-oriented platform is still not popular. By researching the reason, we find it‘s not because the concept of agent does not apply to CS, but many agent-oriented platforms are lack of usability and dynamic. For that, we propose Jadexx which is extend by Jade and hope that agent-oriented programing will be more popular.

In the design of agent deliberation, Jadexx uses four designs in order to make usability and dynamic better. The four designs are 1) Using Java. It makes user which is familiar with Java easily to use it. Moreover Java is common now. 2) Implement the desire and Intention storing structure. It will make Jadexx easy to use when the user is familiar with BDI concept. 3) Add a new plan called meta plan which is mechanism to implement meta-level reasoning. 4) Define an adjustable priority for each desire. It makes agent select different plans in different situations.

Finally, we implement a Jadexx example and compare the usability and dynamic with Jason and Jadex. We find Jadexx is better in the two functionalities.

[1] A. Ricci. et. al., “Agent-oriented Computing: Agents as a Paradigm for Computer Programming and Software Development,” Future Computing 2011.
[2] Telecom Italia Lab, ”Jade” [online]. Available: http://jade.tilab.com/index.html.
[3] IEEE Foundation for Intelligent Physical Agents, “FIPA” [online]. Available: http://www.fipa.org/.
[4] Wikipedia, “Belief¬-desire-Intention software model,” [online]. Available: http://en.wikipedia.org/wiki/Belief%E2%80%93desire%E2%80%93intention_software_model.
[5] Active Component, ”Jadex” [online]. Available: http://www.activecomponents.org/bin/view/About/Features.
[6] A. Pokahr, L. Braubach and W. Lamersdorf. Jadex: A BDI reasoning engine. In J. D. R. Bordini, M. Dastani and A. E. F. Seghrouchni, editors, Multi-Agent Programming, pages 149--174. Springer Science+Business Media Inc., USA, 9 2005. Book chapter.
[7] R.H. Bordini, J.F. Hubner, “BDI Agent Programming in AgentSpeak Using Jason,” Department of Computer Science, University of Durgam，2006.
[8] Jen-Yen Chen, Towards an Agent-oriented Programming Paradigm, Department of Computer Science and Information Engineering, National Central University, 2013.
[9] 林廷原, 使用JADEX開發代理人系統 , 中央大學資工系碩士論文, 2012.
[10] W3C OWL Working Group OWL Web Ontology Language Guide, [online]. Available: http://www.w3.org/TR/owl-guide#owl.