NVIDIA DRIVE™ Infrastructure encompasses the complete data center hardware, software, and workflows needed to develop and validate autonomous driving technology—from raw data collection through validation. It provides the building blocks required for neural network development, training and validation, replay, and testing in simulation.
Building autonomous vehicles requires massive amounts of data. Managing and curating this data takes high-performance compute, as well as intelligent training methods. NVIDIA® DGX™ Systems and advanced training tools enable streamlined, large-scale training and optimization of deep neural networks (DNNs). Using the power of GPUs and AI, developers can comprehensively train DNNs for autonomous vehicle perception, planning, driving, and more.
It’s impossible for an autonomous vehicle to encounter every possible traffic situation while testing on public roads. In simulation, virtual vehicle fleets can drive millions of miles across a broad range of scenarios—from routine driving to rare or even dangerous situations—with greater efficiency, cost-effectiveness, and safety than in the real world. The DRIVE Constellation™ simulation platform comprises two side-by-side servers that generate the sensor output from the virtual car and streams that data into the DRIVE AGX AI car computer running the AV stack to make real-time decisions. Vehicle control commands are then sent back to the simulator. This closed-loop process enables bit-accurate, timing-accurate, hardware-in-the-loop testing.
As autonomous driving software develops and improves, it’s vital that new versions can be tested against previously captured sensor data to avoid regression. With the DRIVE Constellation hardware-in-the-loop platform, developers can replay driving data and compare the performance of the latest self-driving system to past versions. When combined with simulation testing, the DRIVE Constellation platform provides a comprehensive solution to cloud-based validation of autonomous driving technology.
Find out how you can start developing AI-powered autonomous vehicles.