A new ADMORPH paper has been accepted for publication. The paper, entitled “Stability of Linear Systems under Extended
Weakly-Hard Constraints”, will appear in IEEE Control Systems Letter, a highly ranked journal (Scimago Q1).
The paper proposes a comprehensive stability analysis for control systems subject to deadline misses bounded by the weakly-hard model. In the paper, the classical weakly-hard model is extended in order for the analysis to cover additional scheduler configurations. Using properties inherent to the weakly-hard model, we also prove analytic bounds on the stability of embedded systems subject to a broad class of other weakly-hard constraints. The proposed analysis calls for modularity and separation of concern, thus bringing the assessment of control systems stability one step closer to the real-time implementation.
The ADMORPH project was present at the HiPEAC 2022 Conference, which took place in Budapest, Hungary, from 20 to 22 of June.
ADMORPH posters were visible in the coffee-break area throughout the entire event, while an ADMORPH flyer highlighting the project vision, objective, use cases and architecture and technologies, was distributed to HiPEAC participants.
Furthermore, on June 20 the project participated in the DL4IoT workshop, with a presentation given by Stafanos Skalistis.
Marine Kadar works as project research engineer at SYSGO. She received her PhD from the Real-Time Systems Chair of TU Kaiserslautern in 2022. Her PhD study applied in the scope of FORA, a European training network in Horizon 2020 Research and Innovation project. She investigated how to develop and deploy intrusion detection solutions into embedded mixed-criticality systems, evaluating the solutions in an industrial environment, using SYSGO’s PikeOS real-time hypervisor. Before her PhD, Marine received her engineering degree from ENSIMAG engineering school (France), with a specialization on embedded systems.
Was there a particular reason for you to take an engineering course?
Retrospectively, taking an engineering course was a logical decision in my education path. At school, I have always been interested in science and math. Most of all, I find scientific reasoning fascinating and I enjoy solving practical problems.
After 2 years of generic preparatory classes, I decided to take a course in math and computer science. I then focused on embedded systems, at system low level with hardware mechanisms and operating system programming. I particularly liked security-related problems: e.g. attack methods to hijack a program and countermeasures to protect a systems against such threats. I chose to do my PhD in the domain of embedded system security with the motivation to learn more on the topic.
Since you got your PhD, not too long ago, you already worked in a few companies. Was your decision to go to the industry motivated by the will to do more applied work, rather than fundamental research?
During my engineering studies, I had the opportunity to do internships in a research lab and several companies. I enjoyed different aspects of both work environments: for example, real-life problems in industry and relative freedom in research. That is one main reason, why I chose to do a PhD in an industrial environment. The principal contribution of my PhD was indeed to integrate and evaluate security solutions into industrial embedded mixed-criticality systems, using a commercial industrial platform.
The STEM areas are typically male dominated. Did you felt any difference concerning the opportunities given to women, when moving from an academic environment to the industry?
I did not feel a difference between my studies in the university and working in the industry in regard to opportunities given to women. As there were very few women in my computer science course, I logically met very few women with programming skills in the companies where I worked.
Can you tell us about what you are working on in the context of ADMORPH?
In ADMORPH, I am responsible for SYSGO’s contributions in the project. I have been working on developing PikeOS (i.e. SYSGO’s real-time hypervisor) extensions to support runtime mechanisms in ADMORPH system architecture. These include fault detection methods and adaptation solutions to support ADMORPH use-cases.
My main focus is on the topic of fault and intrusion detection, which is directly related to my PhD work. I contributed to the implementation of a hardware-assisted framework based on ARM CoreSight processor tracing technology to implement transparent control-flow monitoring of a user-level application during runtime. I participated to the evaluation of this monitoring framework in a real-life environment using PikeOS real-time hypervisor and a commercial hardware target.
Finally, if you would be talking to young girls about their future career, would you try to convince them about the relevance of working in a technological area? How would you motivate them?
Since I started to study general science and then when I specialized in computer science, the ratio of girls fell down below 10%. In my opinion, this result is mostly due to personal choices (without constraint), which are influenced by our day-to-day environment: family, role models, school, medias, politics, etc. Changing the society towards more gender equality is a non-trivial political goal. I do not think that one speech can easily make someone change his/her mind on such personal matter.
Hence, more than convincing young girls to choose a career in a technological area, I would emphasize the need for them to do what they like and to be independent, so that they can be free to make their own choices.
Thanks a lot for your answers, Marine!
Finally, after two years of virtual meetings due to the COVID-19 pandemic, the #ADMORPH project partners were able to physically meet once again! The consortium meeting took place at the Faculty of Sciences of the University of Lisbon, in Portugal, from the 11th to 13th of May.
During the meeting it was possible to review the work that is being done in all work packages, collect important feedback from our Advisory Board members (thank you!!!), and work together to ensure a nice integration of the work around the defined use cases in the naval, railway and avionics domains.
Not less important, we also used the opportunity to socialize and get to better know each other after work, enjoying together some of the nice things that Lisbon has to offer: food, views and weather 🙂
Computation is becoming cheaper and cheaper, with a lot of dedicated core capacity. Because of this, we are plugging in components in our control systems, like anomaly detector and predictive maintenance algorithms, that can help us take better advantage of the computational power for something useful. However, this additional load may come at the cost of problems and bugs, that manifest themselves as deadline misses. The controller that should regulate the plant does not manage to compute a fresh control signal on time.
The paper “Stability and Performance Analysis of Control Systems Subject to Bursts of Deadline Misses“, that received the best paper award at ECRTS 2021 sets off to try to answer the question: should we worry about this? Control systems are designed to be robust to a large set of disturbances, ranging from noise to unmodelled dynamics. Are computational delays and faults a problem?
Recent work on the weakly hard model – applied to controllers – has shown that control tasks can also be inherently robust to deadline misses. However, existing exact analyses are limited to the stability of the closed-loop system. In the paper, we show that stability is important but cannot be the only factor to determine whether the behaviour of a system is acceptable also under deadline misses. We focus on systems that experience bursts of deadline misses and on their recovery to normal operation. We apply the resulting comprehensive analysis (that includes both stability and performance) to a Furuta pendulum, comparing simulated data and data obtained with the real plant.
We further evaluate our analysis using a benchmark set composed of 133 systems, which is considered representative of industrial control plants. Our results show the handling of the control signal is an extremely important factor in the performance degradation that the controller experiences, a clear indication that only a stability test does not give enough indication about the robustness to deadline misses.