Nvidia is bringing its drive av software with enhanced level 2 point to point driver assistance capabilities to u.s. roads by the end of this year, starting with mercedes benz and its all new cla model. The cla is the first mercedes benz vehicle to feature the mb.os platform, and it uses nvidia’s full stack drive av software, artificial intelligence infrastructure and accelerated compute to power advanced driver assistance. The design is intended to support over the air updates for future upgrades and new features, including planned enhancements to mb.drive assist pro that may be available ex factory and through the mercedes benz store.
The mercedes benz cla recently received a five star european new car assessment program euroncap safety rating, with mb.drive active safety features contributing to accident mitigation, avoidance and the overall top score. Nvidia describes drive av as a dual stack architecture that combines an artificial intelligence end to end core driving stack with a parallel classical safety stack built on the nvidia halos safety system, adding redundancy and safety guardrails. This unified architecture enables advanced level 2 automated driving functions such as point to point urban navigation in complex city environments, proactive collision avoidance, automated parking in tight spaces and cooperative steering between the system and the driver.
Nvidia deep learning models power artificial intelligence assisted urban driving that can interpret traffic holistically to handle lane selection, turns and route following in congested or unfamiliar areas, as well as proactively respond to vulnerable road users like pedestrians, cyclists and scooter riders. Beyond in vehicle capabilities, nvidia and mercedes benz are using nvidia omniverse libraries and digital twins of factories and assembly lines to design, plan and optimize manufacturing operations virtually, while omniverse and the nvidia cosmos platform are used to test and validate intelligent driving software in simulation before real world deployment. A cloud to car development pipeline ties together nvidia dgx training infrastructure, omniverse and cosmos simulation for billions of virtual miles, and in vehicle drive agx compute with the drive hyperion architecture, creating a closed loop system intended to accelerate algorithm iteration, improve accuracy, validate safety in rare edge cases and support scalable deployment across multiple vehicle platforms, including those of other global automaker partners.
