#	Pagina
attuale pagina	/open-h2020/projects/198500/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - oCPS (Platform-aware Model-driven Optimization of Cyber-Physical Systems)

Teaser

Summary

Many modern industrial systems fall in the realm of Cyber-Physical Systems (CPS) because of the tight interaction between computation, communication and control elements (the cyber part), and physical processes (the physical part) within these systems. Requirements related to cost, quality and reliability enforce designs with over-provisioning of platform resources (computation, communication, memory) by large margins at each phase to be able to fulfil system-level requirements in the worst-case scenarios. To replace such overly conservative design process, there is an urgent need for integrative design trajectories that allow for tradeoffs between cost, quality and reliability coping with the tight coordination between the cyber and the physical components. This gives rise to the need for models that accurately capture the interaction between various components (e.g., software, electronics, mechanics, algorithms, power, energy, etc.) and novel design methods that exploit the artefacts of the underlying platforms.
The key scientific objective of the oCPS program is to enable the design of a new generation of cost-effective, quality-driven and reliable CPS by developing model-driven design methods that capture the interaction between different models at various design layers, that take into account physical constraints and processes, and that introduce platform-awareness at all levels. The program aims to train a generation of young researchers in cross-disciplinary thinking and deliver industrially validated tool chains.
The consortium is composed by nine beneficiaries (Eindhoven University of Technology, Philips Healthcare, Technical University of Munich, Inchron, Fortiss, TU Dortmund, Royal Institute of Technology in Stockholm, IMSYS and Odys) involving 4 European countries (Netherlands, Germany, Sweden and Italy). Besides that, 8 partners (Technolution, TNO, Scania, Ericsson, Siemens, University of Ulm, IMT Lucca, and TU Vienna) support the consortium with training, secondments, use cases and tool chains.

Work performed

oCPS is well on track, with a good team of early stage researchers (ESR) and staff, and a sound basis for research and innovation in the CPS domains. ESR research is driven by industrial use cases. Solutions are integrated in a number of toolchains. Five industrial drivers use cases have been identified to capture major research challenges to be addressed in oCPS: platooning, the smart car, the smart grid, smart imaging, and healthcare equipment. The methods are being validated in a number of toolchains, bringing together multiple relevant tools for a specific application domain.
The following results have been obtained by the ESRs in their individual projects so far:
ESR1: A design flow for reducing overall cost, improving resource efficiency (and hence performance and power dissipation), and improving Quality-of-Control in data-intensive multiprocessor controllers for CPS has been developed.
ESR2: An event triggered deterministic threshold algorithm and a controller based on the well-known LQG controller have been created, aiming to integrate control and communication design so that communications and power needed by applications is reduced.
ESR3: A networked supervisory control system for hybrid systems has been developed that ensures high-level control requirements for a networked CPS.
ESR4: To meet the low energy requirements of modern day computing platforms, a framework for quality controlled approximate synthesis of combinational circuits has been developed.
ESR5: Parameterized models describing hardware dependencies and performance characteristics of embedded software have been developed using Domain-Specific Language technology to accurately predict CPS performance.
ESR6: A two-layer hierarchical control architecture for vehicle platooning has been developed taking into account the communication standards considered by OEMs, with the goal of safer driving, fuel reduction and increase of the road capacity.
ESR7: Conditions under which compositional abstractions of networks of stochastic hybrid systems can be constructed have been derived. These techniques help to achieve societal goals of developing high-quality and reliable CPS in a cost-effective way.
ESR8: Architectural guidelines and sufficient conditions for ergodicity of physically coupled networked control systems (NCS) have been developed, to guarantee efficiency, scalability and flexibility of the network in an NCS.
ESR9: An improved modular performance analysis with real-time calculus has been developed combining code-level and model-level level timing analysis to significantly improve the design trajectory for real-time applications in various domains.
ESR10: The synthesis problem for correct-by-construction synthesis of control software for physical systems and processes was formulated and some synthesis strategies were investigated.
ESR11: Approaches to work with partial and heterogeneous models have been studied, aiming to facilitate the work of the CPS designer by enabling more elaborate design activities while at the same time cutting time in the development. A real world CPS design challenge of the Dutch Rijkswaterstaat has been analysed.
ESR12: Traffic models predicting future traffic, calculating admissible heavy-duty vehicle (HDV) speeds, and deriving fuel-optimal control have been developed to significantly improve commercial HDV platooning by reducing fuel consumption and improving road efficiency.
ESR13: Partitioning methods in multi-view modelling have been analysed, aiming to increase CPS design process efficiency through a complexity reduction of the design process.
ESR14: Self-awareness principles are being developed for energy efficient, cost constrained Internet of Things devices with a high Quality of Service. Models and principles have been partially implemented in chronSIM from INCHRON and RoSA from TU Vienna.
ESR15: To bridge the gap between state-of-the-art model predictive control (MPC) theory and industrial practice for MP

Final results

The key innovation of oCPS is the development of multidisciplinary, cross-layer model-driven design methods. These are realized in a number of domain-specific and general toolchains. At the end of the project, it is expected to have a set of toolchains for research and industrial use.
The oCPS activities have the following impact: human resource development for next generation CPS design, employability and career opportunities for the ESRs, multi-disciplinary research training at the doctoral level, stronger industry-academia connections, aligned R&D agendas, including training on the job, entrepreneurship and start-ups, and efficient design trajectories, shorter time-to-market.
Seven scientific publications have been accepted so far. We expect to increase that number to 40 publications at the end of the project.