A survey of Model Driven Engineering in robotics

Abstract Robots are complex to develop due to the diversity of hardware, software components and the absence of common standards. To deal with these problems, Model-Driven Engineering (MDE) is used to systematize the software development. This work aims to provide a comprehensive overview of existing model-based approaches in robotics. We present a classification of the 63 papers selected comparing past expectations with present achievements, and showing that the self-adaptation of robots has been poorly explored. With advances in machine learning, there has been an increase in the possibilities of using model-based approaches aiming at greater autonomy of robotic software.