The European Union Aviation Safety Agency (EASA) had issued a call for tenders for the Machine Learning Application Approval (MLEAP) research project funded by the Horizon Europe research and innovation program. It selected APSYS, an Airbus subsidiary, to implement the MLEAP project in partnership with LNE & NUMALIS.
Airbus decided last July to merge the services activities of Airbus Cybersecurity and Apsys, specialized in security and industrial risk management, and created Airbus Protect, which is therefore the project leader for the next two years.
The Laboratoire National de Métrologie et d’Essais (LNE) and Numalis, a French company that publishes innovative software providing tools and services to make AI reliable and explainable, have been working with Airbus Protect on the project since May 2022. The project focuses on the approval of machine learning (ML) technology for systems to be used in safety-related applications in all areas covered by the EASA Basic Regulation and is funded by the Horizon Europe program with €1,475,400.
EASA has been interested in potential applications based on machine learning and deep learning in safety-critical applications for several years, and in fact published its artificial intelligence roadmap in February 2020, followed by a concept paper: ” First usable guidance for level 1 machine learning applications” in April 2021. This concept paper presents a first set of objectives for level 1 AI (human assistance), in order to anticipate future EASA guidance and requirements for safety-related machine learning (ML) applications.
Project objectives
The partners will focus on streamlining certification and approval processes by identifying concrete ways to comply with the learning assurance objectives of the EASA guidelines for AI/ML applications (Levels 1, 2, and 3 as defined in the EASA AI Roadmap), with a focus on Levels 1B (enhanced human assistance) and 2 (human/machine collaboration). At level 3, the machine becomes more autonomous.
The medium-term effect of the project will be to alleviate some of the remaining restrictions on the acceptance of ML applications in safety-critical applications.
Expected results
The research results will consist of a set of reports identifying a set of methods and tools to address the following three major topics:
- Guarantees on the “generalization of the machine learning model
- Guarantees on the “completeness and representativeness of the data
- Guarantees on the robustness of the algorithm and the model
In parallel with the project, at least one full-scale aviation use case will need to be developed to demonstrate the effectiveness and usability of the proposed methods and tools.
The work will be broken down as follows:
- Task 1: Methods and tools for assessing completeness and representativeness of datasets (training, validation and testing) in data-driven ML and distance learning;
- Task 2: Methods and tools for quantifying generalization guarantees for ML and DL models;
- Task 3: Methods and tools for verification of an ML algorithm and model robustness/stability;
- Task 4: Communication, dissemination, knowledge sharing, stakeholder management;
- Task 5: Project Management.
Translated from L’EASA lance le projet de recherche « Machine Learning Application Approval » (MLEAP)
