Computer‐based systems are now expected to evolve during their service life to cope with changes of various nature, ranging from evolution of user needs, eg, additional features requested by users, to system configuration changes, eg, modifications in available hardware resources. When considering resilient embedded systems that must comply with stringent dependability requirements, the challenge is even greater, as evolution must not impair dependability attributes. Maintaining dependability properties when facing changes is, indeed, the exact definition of resilient computing. In this paper, we consider the evolution of systems with respect to their dependability mechanisms and show how such mechanisms can evolve with the system evolution, in the case of ROS, the robot operating system. We provide a synthesis of the concepts required for resilient computing using a component‐based approach. We particularly emphasize the process and the techniques needed to implement an adaptation layer for fault tolerance mechanisms. In the light of this analysis, we address the implementation of adaptive fault tolerance on ROS in 2 steps: Firstly, we provide an architecture to implement fault tolerance mechanisms in ROS, and secondly, we describe the actual adaptation of fault tolerance mechanisms in ROS. Beyond the implementation details given in the paper, we draw the lessons learned from this work and discuss the limits of this run‐time support to implement adaptive fault tolerance features in embedded systems.

International audienceSystems are expected to evolve during their service life in order to cope with changes of various natures, ranging from fluctuations in available resources to additional features requested by users. For dependable embedded systems, the challenge is even greater, as evolution must not impair dependability attributes. Resilient computing implies maintaining dependability properties when facing changes. Resilience encompasses several aspects, among which evolvability, i.e., the capacity of a system to evolve during its service life. In this paper, we discuss the evolution of systems with respect to their dependability mechanisms, and show how such mechanisms can evolve accordingly. From a component-based approach that enables to clarify the concepts, the process and the techniques to be used to address resilient computing, in particular regarding the adaptation of fault tolerance (or safety) mechanisms, we show how Adaptive Fault Tolerance (AFT) can be implemented with ROS. Beyond implementation, we draw the lessons learned from this work and discuss the limits of this runtime support to implement such resilient computing features in embedded systems

The use of over-the-air updates has attracted very much interest these last few years with the software-intensive development of embedded systems in the car industry. The development of autonomous driving and ADAS (Advanced Driver Assistance Systems) renders over-the-air updates mandatory, for both user satisfaction and economic reasons. How to make sure that remote updates of critical ADAS do not have an impact on safety? This is the question we tackle in our work with a major car manufacturer. This paper is a progress report. We summarize our approach involving AFT (Adaptive Fault Tolerance) implemented on ROS (Robot Operating System), describe the simulation platform we have developed to experiment and validate over-the-air updates of ADAS and AFT, and finally draw some lessons learnt and perspectives. I.