On the Testability of BDI Agent Systems (Extended Abstract)
M. Winikoff, and S. Cranefield. Proceedings of the International Joint Conference on Artificial Intelligence, page 4217-4221. AAAI Press, (2015)
Abstract
Before deploying a software system we need to assure ourselves (and stakeholders) that the system will behave correctly. This assurance is usually done by testing the system. However, it is intuitively obvious that adaptive systems, including agent-based systems, can exhibit complex behaviour, and are thus harder to test. In this paper we examine this öbvious intuition" in the case of Belief-Desire-Intention (BDI) agents, by analysing the number of paths through BDI goal-plan trees. Our analysis confirms quantitatively that BDI agents are hard to test, sheds light on the role of different parameters, and highlights the enormous difference made by failure handling.
%0 Conference Paper
%1 IJCAI1510717
%A Winikoff, Michael
%A Cranefield, Stephen
%B Proceedings of the International Joint Conference on Artificial Intelligence
%D 2015
%I AAAI Press
%K agents bdi coordination goals group middleware myown zookeeper
%P 4217-4221
%T On the Testability of BDI Agent Systems (Extended Abstract)
%U http://www.aaai.org/ocs/index.php/IJCAI/IJCAI15/paper/view/10717
%X Before deploying a software system we need to assure ourselves (and stakeholders) that the system will behave correctly. This assurance is usually done by testing the system. However, it is intuitively obvious that adaptive systems, including agent-based systems, can exhibit complex behaviour, and are thus harder to test. In this paper we examine this öbvious intuition" in the case of Belief-Desire-Intention (BDI) agents, by analysing the number of paths through BDI goal-plan trees. Our analysis confirms quantitatively that BDI agents are hard to test, sheds light on the role of different parameters, and highlights the enormous difference made by failure handling.
@inproceedings{IJCAI1510717,
abstract = {Before deploying a software system we need to assure ourselves (and stakeholders) that the system will behave correctly. This assurance is usually done by testing the system. However, it is intuitively obvious that adaptive systems, including agent-based systems, can exhibit complex behaviour, and are thus harder to test. In this paper we examine this "obvious intuition" in the case of Belief-Desire-Intention (BDI) agents, by analysing the number of paths through BDI goal-plan trees. Our analysis confirms quantitatively that BDI agents are hard to test, sheds light on the role of different parameters, and highlights the enormous difference made by failure handling.},
added-at = {2016-05-24T05:29:34.000+0200},
author = {Winikoff, Michael and Cranefield, Stephen},
biburl = {https://www.bibsonomy.org/bibtex/247abf31bcaacb4bb31fd1a862e13afc9/scranefield},
booktitle = {Proceedings of the International Joint Conference on Artificial Intelligence},
description = {Winikoff},
interhash = {c0edbb6ddb33f2cb37e48eb5122e9c09},
intrahash = {47abf31bcaacb4bb31fd1a862e13afc9},
keywords = {agents bdi coordination goals group middleware myown zookeeper},
pages = {4217-4221},
publisher = {AAAI Press},
timestamp = {2016-05-24T05:29:34.000+0200},
title = {On the Testability of BDI Agent Systems (Extended Abstract)},
url = {http://www.aaai.org/ocs/index.php/IJCAI/IJCAI15/paper/view/10717},
year = 2015
}