Optimum testing of control software for engine management systems

The paper describes an effective method for control software testing. Starting from a previous release, the authors propose a proper update of their testing methodology to make it feasible for a new class of software. More in depth, the testing methodology was designed to generate a set of input combinations (so-called test cases), each one accounting for the interdependence in terms of amplitude among the software inputs. In this paper, the testing methodology has been significantly updated to take into account also the time interdependence among the software inputs. The technique has been applied to a typical automotive case study that concerns with the validation of monitoring and control software for the engine management system.