Automating the product derivation process of multi-agent systems product lines

Agent-oriented software engineering and software product lines are two promising software engineering techniques. Recent research work has been exploring their integration, namely multi-agent systems product lines (MAS-PLs), to promote reuse and variability management in the context of complex software systems. However, current product derivation approaches do not provide specific mechanisms to deal with MAS-PLs. This is essential because they typically encompass several concerns (e.g., trust, coordination, transaction, state persistence) that are constructed on the basis of heterogeneous technologies (e.g., object-oriented frameworks and platforms). In this paper, we propose the use of multi-level models to support the configuration knowledge specification and automatic product derivation of MAS-PLs. Our approach provides an agent-specific architecture model that uses abstractions and instantiation rules that are relevant to this application domain. In order to evaluate the feasibility and effectiveness of the proposed approach, we have implemented it as an extension of an existing product derivation tool, called GenArch. The approach has also been evaluated through the automatic instantiation of two MAS-PLs, demonstrating its potential and benefits to product derivation and configuration knowledge specification.

Evolutionary Testing of Autonomous Software Agents

A system built in terms of autonomous agents may require even greater correctness assurance than one which is merely reacting to the immediate control of its users. Agents make substantial decisions for themselves, so thorough testing is an important consideration. However, autonomy also makes testing harder; by their nature, autonomous agents may react in different ways to the same inputs over time, because, for instance they have changeable goals and knowledge. For this reason, we argue that testing of autonomous agents requires a procedure that caters for a wide range of test case contexts, and that can search for the most demanding of these test cases, even when they are not apparent to the agents’ developers. In this paper, we address this problem, introducing and evaluating an approach to testing autonomous agents that uses evolutionary optimization to generate demanding test cases.