Nvidia Has Announced Its Self Driving Toolkit Hyperion 8 For 2024 Vehicles

As carmaker's are increasingly aiming to roll out automated driving features with the proliferation of robotaxis and self-driving vehicles - Nvidia has launched its latest set of compute and sensor toolkits. Nvidia founder and CEO Jensen Huang revealed its Hyperion 8 stack at its GTC event. Hyperion 8 is a platform that includes sensors, compute and software that's needed for the 2024 models.

Hyperion 8 is the latest iteration of end to end Nvidia Drive platform that automakers can customise as per their needs. Hyperion 8 was first announced on April 8 supplying 12 cameras, nine radars, and one LiDAR which is part of the platform. Nvidia has tied up with Luminar to deliver the LiDAR and Continental, Sony and Valero are supplying the sensors for Hyperion 8.

"The opportunity here is that as Nvidia's system gets designed in (to production vehicles), we also ultimately get designed in," said Luminar founder and CEO Austin Russell.

Hyperion 8 is basically a turnkey solution that Nvidia sells to automakers dabbling in self-driving technology. It enables core compute, middleware, and AI functions on top. The announcement also ties in with Nvidia's acquisition of DeepMap, an AI-based platform that can replicate real-time maps with great precision and realism.

Porsche And Nvidia Working On Autonomous Vehicle AI

It also acts like a personal concierge product that provides communication between automated driving systems, the passengers and can handle parking as well. All of this now fits into the Hyperion 8 stack.

Nvidia already counts Cruise, Mercedes-Benz, Volvo, TuSimple, and Amazon-owned Zoox as its customers. Now it is adding Lotus, Qcraft, and Weltmeister as clients.

"We do not expect to have a 100% market share, but we have a dominant portion of the market that's developing on Nvidia Drive," Danny Shapiro, vice president of automotive at Nvidia said in a briefing ahead of GTC.

"And the reason that is, is because we have this entire end-to-end solution. It's not just about one thing that goes in the car but it's about the data center, it's about the simulation, it's about the vehicle, and having that on the same architecture is a huge advantage," he added.