In the context of validation and Hardware-in-the-Loop (HiL) test benches, the evaluation of driving function tests currently represents a major source of manual workload for employees. The project addresses this challenge by developing a modular pattern recognition system for both image and audio data. This system will enable the economic and efficient automation of suitable test evaluations.
The methodological foundation is Systems Engineering, which will be applied to define requirements and design specifications. Based on these, an economic assessment of different tests will be conducted to determine their suitability for automation. Subsequently, modular test bench concepts will be developed to support automated evaluation of identified test spectra. These concepts will then be combined with economic assessments, allowing for strategic planning and roll-out of automation measures.
To realize these concepts, the project leverages advanced AI-based methods, including computer vision for image recognition and large language model (LLM) technology for audio analysis. The integration of these technologies aims to streamline evaluation processes, minimize human error, and support standardization across multiple test domains.
The overarching goal of the project is to significantly reduce manual effort in test evaluation through the deployment of modular, AI-driven automation systems. Specifically, the project seeks to:
