Skip to main content

Platform-aware Model-driven Optimization of Cyber-Physical Systems

Article Category

Article available in the folowing languages:

New customisable toolchains turn cyber-physical systems green

Cyber-physical systems are playing an important role in the transition towards a sustainable economy. The oCPS project is making sure they also comply with near-zero power consumption requirements.

Digital Economy

Cyber-physical systems (CPSs) are key infrastructures for our modern society. All around us, computers, machines and humans are working together using embedded computers and networks. Almost everything in modern society is controlled by embedded computers, from miniscule pacemakers to modern cars, which are packed with hundreds of millions of lines of code, and from automated warehouses to almost all devices encompassed by the Internet of things (IoT). Building on Europe’s strong market position in electronics, microsystems and embedded systems, the future of CPSs is a key element in efforts to realise a sustainable economy. It is just as important to ensure these systems adhere to requirements for near-zero power consumption. Working in this direction, the EU-funded oCPS project developed a multidisciplinary training programme involving computer science, electrical, mechanical, control and software engineering.

Reducing energy consumption

“One challenge is to make CPS energy-efficient,” notes Dip Goswami, oCPS project investigator. “As stated by the European Commission’s Green Deal priorities, to have an energy-neutral Europe by 2050 is of paramount importance to reduce the effects of climate change on nature, citizens and business. By developing techniques to improve energy efficiency of systems, oCPS contributed to that objective. In addition, embedded computers need to be safe and cost-effective for a wider acceptance in our society.” For instance, the energy efficiency of a mobile telephone depends on its battery and the apps it uses, as well as the embedded computing architecture. The oCPS project developed model-driven design methods to deal with the multiple layers. “A mobile camera, for example, does not necessarily need high-resolution images for many common applications like face recognition,” explains Goswami. “An approximate image may suffice, saving more than 80 % on energy consumption.” The project developed customisable toolchains composed of tooling solutions of different relevant layers involving hardware, software and physics. For instance, a truck platoon solution may be tested in an integrated tooling environment that can accurately model traffic behaviour, vehicle-to-vehicle communication and controlling algorithms. By integrating the developed design methods in six customisable toolchains targeting varied domains (such as autonomous vehicles, platooning, imaging, multi-agent systems and healthcare), the project showed improvement in terms of energy efficiency, cost and safety. “For example, we have shown above 80 % energy improvement for vision-based applications in cars,” notes Goswami. “Similarly, a fuel saving of 7 % is shown by our platooning solution, which is substantial given the number of trucks on our roads.”

Working across disciplines

According to Goswami, while working across disciplines was challenging, it led to interesting results. The project, undertaken with the support of the Marie Skłodowska-Curie programme, was also able to forge collaborations among experts and researchers with different expertise through industrial use cases who came through several secondments as well as network-wide meetings and regular discussion sessions. This led to multidisciplinary methods and the customisable toolchains. “We envision that the developed methods will be further integrated in industrial design trajectories,” says Goswami. “Naturally, this requires a step forward towards a higher level of maturity. Further, the methods imply a paradigm shift that needs to be embedded in our education trajectories, courses and student projects.”


oCPS, toolchains, computers, cyber-physical systems, CPS, embedded computers, energy efficiency, Green Deal

Discover other articles in the same domain of application