Making critical autonomous AI-based systems safe

Making critical autonomous AI-based systems safe

Objectives

To improve the explainability and traceability of DL components

To provide clear safety patterns for the incremental adoption of DL software in Critical Autonomous AI-based Systems (CAIS)

To integrate the SAFEXPLAIN libraries with an industrial system-testing toolset

To create architectures of DL components with quantifiable and controllable confidence, and that have the ability to identify when predictions should not be released based on applicability’s scope or security concerns

To design, implement, or update selected representative DL software libraries according to safety patterns and safety lifecycle considerations, meeting specific performance requirements on  relevant platforms

Deep Learning (DL) techniques are key for most future advanced
software functions in Critical Autonomous AI-based Systems (CAIS) in
cars, trains and satellites. Hence, those CAIS industries depend on their
ability to design, implement, qualify, and certify DL-based software
products under bounded effort/cost

Case studies

Railway: This case studies the viability of a safety architectural pattern for the completely autonomous operation of trains (Automatic Train Operation, ATO) using intelligent Deep Learning (DL)-based solutions.

Space: This case employs state-of-the-art mission autonomy and artificial intelligence technologies to enable fully autonomous operations during space missions. These technologies are developed through high safety-critical scenarios.

Automotive: This case develops advanced methods and procedures that enable self-driving cars to accurately detect road users, estimate their distance from the vehicle, and predict their trajectories while adhering to both safety and explainability requirements.

Consortium sets course for last year at Barcelona F2F

Consortium sets course for last year at Barcelona F2F

Members of the SAFEXPLAIN consortium met in Barcelona, Spain on 29-30 October 2024 to discuss the project's process at the end of the second year of the project. With one year to go, project partners used this in-person meeting to close loose ends and ensure that...

Second IAB Meeting Confirms SAFEXPLAIN Advancements at Start of Year 3

Second IAB Meeting Confirms SAFEXPLAIN Advancements at Start of Year 3

The SAFEXPLAIN project met with members of its industrial advisory board on 03 October 2024 to present project advancements at the beginning of the project’s third and final year. This meeting was important for ensuring the project’s research outcomes align with real-world industry needs.

RISE explains XAI for systems with Functional Safety Requirements 

RISE explains XAI for systems with Functional Safety Requirements 

The SAFEXPLAIN project is analysing how DL can be made dependable, i.e., functionally assured in critical systems like cars, trains and satellites. Together with other consortium members, RISE has been working on establishing principles for ensuring that DL components, together with required explainable AI supports, comply with the guidelines set forth by AI-FSM and the safety pattern(s).

European Convergence Summit-Digital Booth ADR Exhibition

European Convergence Summit-Digital Booth ADR Exhibition

SAFEXPLAIN will have a digital booth as part of the ADR Digital Exhibition, co-located within the European Convergence Summit 2024. This digital booth will showcase the work conducted as part of the SAFEXPLAIN project, including videos, publications, and presentations...

TrustworthyAI Cluster Webinar hosted by ADRA-e

TrustworthyAI Cluster Webinar hosted by ADRA-e

SAFEXPLAIN partner Enric Mezzetti from Barcelona Supercomputing Center will join the ADRA-e hosted webinar on "Trustworthy AI: Landscaping veriable robustness and transparency" on 29 May 2024 from 10-12h. The TrustworthyAI Cluster, nine EU-projects under call Horizon...