FASTER-ΑΙ research project
The FASTER AI (Fully Autonomous Safety- and Time-critical Embedded Realization of Artificial Intelligence) project aims to embed machine learning (ML) inference capabilities within critical hardware infrastructure used in telecommunications, airborne systems, and other vehicles. Current ML workflow programming tools focus on standard AI accelerators and overlook non-commodity use, controlled primarily by dominant cloud vendors. However, as ML inference takes over traditional heuristic- and control-based decision-making in the industry, there is a need to re-purpose the hardware towards the use of ML. The FASTER AI approach streamlines the ML integration pipeline around three core activities to achieve multi-stage cross-compilation of critical logic and ML functions without sacrificing safety and service time guarantees. The methodology is effective for current and future hardware architectures used in critical decision-making industries.
Project manager at MDU: Masoud Daneshtalab