SaiNSOR

Trustworthy AI-Based Sensor Technology: Modular Interface Technologies for the Cross-Domain Measurement Systems of the Future

Abstract

The SaiNSOR project – an initiative of the German Aerospace Centre (DLR) led by the DLR Institute of Space Research – aims to develop trustworthy, AI-supported data processing procedures for measurement systems. At the heart of the work is the development of a modular, cross-domain interface technology that integrates multimodal sensor systems, closely links data processing and sensor technology – particularly for safety-critical applications – and sustainably strengthens confidence in data quality through qualitative evaluation systems.
This urgent need stems from the challenges posed by modern AI systems: Whilst their ‘black box’ nature often results in opaque decision-making processes, the strict requirements of the EU AI Act explicitly demand the provision of high-quality, traceable and transparent data. Tailor-made sensor data is a crucial building block in making AI systems reliable and trustworthy.
Existing siloed solutions involve significant development effort. SaiNSOR addresses these challenges through flexible interface technology that enables multimodal sensor data to be generated, processed and annotated with quality metrics. This ensures data quality and improves interoperability between different areas of application.