This work is about an approach for designing control software for mechatronic and robotic machines. As all system parts (control algorithms, software infrastructure, I/O, and machine) influence each other, its total behaviour needs to be taken into account. Therefore, we use appropriate modelling formalisms, namely discrete event / discrete time for the control software and algorithm part, and differential equations for the machine / robotics part. We combine these two types of models in a co-modelling approach, supporting co-operative design of such cyber-physical systems.In this paper, the co-modelling and co-design approach is elaborated. It consists of four major steps, whereby the models in each are verified via simulation or co-simulation. It is a pure model-driven design approach, implying that all final code is generated from these verified models. This allows for better quality of the resulting embedded software.The method is illustrated using a case study. Structuring models and verification by simulation, whereby relevant simulations are set up such that they can be repeated as real experiments, is beneficial for the design work itself, i.e. making it far more effective and less error-prone.

No abstract

No abstract