Using Models@Runtime for Designing Adaptive Robotics Software: an Experience Report

Abstract

Robotic systems are becoming increasingly complex, as their tasks and working environments become ever richer. As a result, there is an urgent need to provide robots with self-awareness and self-adaptation capabilities that allow them to autonomously deal, among other things, with software and hardware failures, changes in the environment, or interactions with other systems. The use of high-level models that can be adapted at run-time by the robot itself, promises to signi cantly boost the applicability and performance of robotic systems. This paper reports our experience in applying the DiVA model-driven adaptive approach to a robotics case study, describing its bene ts and limitations for robotics.