Trustable Electronics - TRUST-E

Abstract:
TRUST-E – Trustable (sensor-driven) Electronics for Automotive, Alternative Mobility and Industrial Applications is a European PENTA Euripides project funded by the national authorities of Belgium, Germany, and Sweden. In this contribution, we will first present the intention of the project and its structure. Subsequently, we will describe our part in the project. The first topic here is video stream processing, where we intend to detect errors using AI/ML. For this, we first have to develop flexible fault injection capabilities, because faulty data are not available. The second topic covers trustable communication, and the third analyzes existing engineering flows, in particular with respect to obsolescence. The final outcome of the project will be an example implementation of AI/ML applications along the value chain for observing changes in the behavior, derivate heath values form the observations and feed these back for improving the trustability. This AI/ML observation independent of the normal intended functionality will allow to significantly increase the trustability of systems.