Verifying Communicating Agents by Model Checking in a Temporal Action Logic

Abstract

Abstract. In this paper we address the problem of specifying and verifying systems of communicating agents in a Dynamic Linear Time Temporal Logic (DLTL). This logic provides a simple formalization of the communicative actions in terms of their effects and preconditions. Furthermore it allows to specify interaction protocols by means of temporal constraints representing permissions and commitments. Agent programs, when known, can be formulated in DLTL as complex actions (regular programs). The paper addresses several kinds of verification problems including the problem of compliance of agents to the protocol, and describes how they can be solved by model checking in DLTL using automata.

Model checking communicative agent-based systems

“…Other proposals adapt some model checking techniques (for example bounded and unbounded model checking) and propose new algorithms for verifying temporal and epistemic properties [4,5,6]. Giordano and her colleagues [3] addressed the problem of specifying and verifying systems of communicating agents in a Dynamic Linear Time Temporal Logic (DLTL).…”

“…Except the work done in [3], all the other proposals on model checking of agentbased systems are based only on temporal and epistemic logics. In this paper, we propose a model checking-based verification of dialogue game protocols using a temporal and dynamic logic.…”

“…In this paper, we propose a model checking-based verification of dialogue game protocols using a temporal and dynamic logic. In contrast to [3], the dynamic aspect of our logic is represented by action formulae and not by strengthening the until operator by indexing it with the regular programs of dynamic logic. Our protocols are specified as actions that agents apply to propositional commitments.…”

“…For instance, the leaf labeled by (<Ch>,s 0 ) is successful since there is no state s i such that s 0 → Ch s i . The leaf labeled by ( ' φ φ ¬ ∨ , s 3,4 ) is successful because it is a positive leaf and s 3,4 ' φ φ ¬ ∨ . Therefore, TS [5] is accepted by ABTA [10].…”

“…Recently, model checking has been used to verify agent-based systems [1,2,3,4]. Verifying these systems is becoming more and more necessary because they are increasingly used in several critical application domains, such as e-commerce, simulation, distributed collaborative systems, etc [5,6,7].…”