The ECSEL-project AIDOaRT plans to develop AI-augmented automation supporting modeling, coding, testing, and monitoring as part of a continuous development in Cyber-Physical Systems (CPSs), as part of a model-based framework.
AIDOaRT aims to impact organizations where continuous deployment and operations management are standard operating procedures. DevOps teams may use the AIDOaRT framework to analyze event streams in real-time and historical data, extract meaningful insights from events for continuous improvement, drive faster deployments and better collaboration, and reduce downtime with proactive detection.
- AI-aided Digital Twin synthesis: passive learning from operations data, active learning from existing artefacts and integration of existing models for both environmental aspects and models of the system under development.
- AI-aided Improved Verification: development of AI methods for selecting a reduced, but relevant subset of tests based on operations data, synthesised Digital Twins , safety and security analysis results and system code. Including runtime monitoring to guide the AI methods through, e.g., inputs for re-inforcement learning.
- AI-aided Maturity Assessment of system models: using AI methods to derive scenarios that help to evaluate the quality of the predictions given by models used during development.
Facts
Project duration: April 2021 – March 2024
Coordination: Mälardalen University (SE)
Budget: € 24,6 Million
Funding: € 7,4 Million (nur EU)
Partner:
- Industry: AVL (AT), Dynatrace (AT), Honeywell Aerospace (CZ), Volvo Construction Equipment (SE)
- SME: Automated Software Testing (AT), INTECS (IT), ClearSy (FR)
- Academia: University of Aquila (IT), Mälardalen University (SE), TU Graz (AT), JKU Linz (AT)
- (et al.)