Abstract. In this paper we focus on the problem of testing agent programs written in BUpL, an executable, high-level modelling agent language. Our approach consists of two main steps. We first define a formal language for the specification of test cases with respect to BUpL. We then implement test cases written in the formal language by means of a general method based on rewrite strategies. Testing an agent program with respect to a given test case corresponds to strategically executing the rewrite theory associated to the agent with respect to the strategy implementing the test case.
Lacramioara Astefanoaei, Frank S. de Boer, M. Birn