We propose a verification method in the field of automotive control systems integrating the concepts of Formal Methods with testing, aiming at efficient and reliable software development. Although Formal Methods are believed to provide the benefits of their rigorous nature and their inherent capability of automation, only limited cases are known where Formal Methods were applied in system and software development, in practice, due to two major difficulties: appropriate abstraction in modeling and scalability in automated reasoning. Focusing on testing on the other hand, there is the difficulty of selecting reasonable set of tests for given verification objectives. In order to overcome these difficulties, our approach is to present verification criteria for testing to appropriately cover the property with the help of the Formal Method concepts. From the consistency with respect to the abstraction level of models between generic property (such as controllability) and underlying assumptions, we derive test coverage that covers the models and the assumptions. Based on a case study using a set of the artifact of a product system, we propose a criteria-driven approach with potential benefits in that we expect to gain the practical efficiency of testing the automotive control systems with the concept of model-checking.