Imagine being able to test an autonomous vehicle in a near-infinite variety of conditions and scenarios—before it even reaches the road. NVIDIA is making it happen, enabling the industry to safely drive billions of miles in virtual reality with the powerful new NVIDIA DRIVE™ Constellation AV simulator.
DRIVE Constellation uses photorealistic simulation to create a safer, more scalable, and more cost-effective way to bring self-driving cars to our roads. It uses the computing horsepower of two different servers to deliver a revolutionary cloud-based computing platform, capable of generating billions of miles of autonomous vehicle testing.
The first server runs DRIVE Sim software to simulate a self-driving vehicle’s sensors—from cameras to lidar and radar. Powerful GPUs generate photoreal data streams that create a wide range of testing environments and scenarios. This means you can easily test rare and difficult conditions: rainstorms, snowstorms, and sharp glare at different times of the day and night, with different road surfaces and surroundings.
The second server contains a powerful DRIVE Pegasus AI car computer that runs the complete, binary compatible autonomous vehicle software stack that operates inside an autonomous vehicle. It processes the simulated data as if it were coming from the sensors of a car actually driving on the road.
Working together, DRIVE Constellation and DRIVE Sim create a “hardware-in-the-loop” system. In this digital feedback loop, simulated sensor data flows into the Pegasus AV computer system 30 times a second, and is processed. Driving commands from Pegasus are sent back in real-time to control the virtual vehicle traveling in the simulated environment to validate the algorithms and software are operating correctly. All designed for maximum flexibility and highest throughput, while eliminating risk.
NVIDIA automotive solutions are available to automakers, tier 1 suppliers, startups, and research institutions shaping the future of transportation.