STRADVISION and aiMotive Demonstrate Integrated ADAS Validation Pipeline

STRADVISION and aiMotive have demonstrated a combined workflow for advanced driver assistance system validation, announced in a press release. The proof of concept connects perception data with neural simulation to transform real world fleet recordings into synthetic environments suitable for large scale testing.

STRADVISION contributed its SVNet perception platform, which interprets and structures driving scenarios from recorded road data. aiMotive applied its World Extractor and aiSim tools to convert these perception derived scenarios into detailed three dimensional environments using Gaussian Splatting techniques. The resulting synthetic sensor data closely matched real world recordings.

The process runs on cloud infrastructure and covers the full pipeline from data ingestion through reconstruction and scenario generation to synthetic data export. aiMotive’s aiSim simulator is ISO 26262 ASIL D certified, allowing validation in a safety qualified environment. The companies state that this integration improves scenario coverage and reduces the gap between field testing and simulation based validation.

The collaboration establishes a scalable method for transforming fleet recordings into simulation ready assets. It also allows automated creation of diverse driving situations, including rare or complex edge cases that are difficult to capture in the real world.