Press Release: Accelerating the development of AD applications for expanding operational design domains.
GeoMate and dSPACE have entered into a partnership to accelerate the development of advanced driver assistance systems and autonomous driving technologies in continuously expanding operational design domains (ODDs).
GeoMate, a company specializing in AI-based mapping, is integrating its OpenDRIVE road network definitions and highly automated, consistent 3D models of the real-world environment into the dSPACE simulation ecosystem. The integration of the models enables developers of autonomous vehicles to quickly carry out extensive virtual test drives in various scenarios based on high-resolution geographical image data and deep-learning technology. The aim of the collaboration is to improve the accuracy, efficiency, and reliability of simulations and validations of autonomous vehicles.
"GeoMate creates compact HD maps for AD and ADAS systems and simulations with an innovative AI-based approach that increases speed, performance and cost efficiency. Our pipeline uses high-resolution geospatial data as a more cost-effective data source to create compact HD maps. Compared to lidar methods, this only costs a fraction," says Dr. Nastaran Saberi, CTO of GeoMate. "GeoMate's compact HD maps are transformed into simulated real-world environments (RealSimE) and serve as a base layer that enables AD systems to effectively use our simulation data and enhance AD capabilities."
Michael Kluge, Product Manager Scenarios at dSPACE, explains: “The seamless integration of GeoMate's detailed, comprehensive maps into our AURELION and ASM simulation solutions adds tremendous value for our customers and expands their ADAS/AD validation and sensor simulation capabilities.”
Verification and validation solutions such as the dSPACE Automotive Simulation Models ASM and the sensor simulation AURELION are supplemented with necessary simulation artifacts through cooperation with partners. In order to effectively validate functions for autonomous driving and sensors, reality-based 3D environment models are required that accurately represent the underlying semantics of the road. Seamless integration of ASAM OpenDRIVE format from GeoMate into dSPACE platforms is one of the main advantages that allows users to work collaboratively with compatible testing and deployment software stacks.
With this partnership, GeoMate and dSPACE want to set new standards in the simulation and validation landscape for autonomous driving technologies and offer tools that make a significant contribution to the efficient and safe development of autonomous vehicles.
GeoMate is integrating its OpenDRIVE road network definitions and highly automated, consistent 3D models of the real-world environment into the dSPACE simulation ecosystem.
More info here.
