In October 2017, NVIDIA introduced a groundbreaking addition to its DrivePX family of autonomous driving computing platforms at the GT Cuppopoe ecosystem conference. This new platform, codenamed "Pegasus," was designed to push the boundaries of self-driving technology. It was expected to be made available to NVIDIA’s autonomous driving R&D partners in the second quarter of 2018.
Pegasus boasts an impressive 320 TOPS (Trillion Operations Per Second) of computing power, which is up to ten times more powerful than its predecessor, the DrivePX2. This performance boost comes from four key processors: two SoCs based on the Volta GPU architecture and two dedicated GPUs optimized for vision processing and deep learning tasks.
To support full autonomy, Pegasus features a high-performance I/O system capable of handling data from multiple sensors, including 16 pairs of radar, camera, and Optima interfaces. It also includes essential automotive interfaces like CAN, FlexRay, and car Ethernet. The platform offers a total memory bandwidth of 1TB, making it ideal for handling complex sensor data in real-time.
NVIDIA emphasized that Pegasus meets the ASIL-D level of the ISO 26262 standard, ensuring robust functional safety for automotive electronics. Redundant design elements are incorporated to guarantee system reliability, enhancing both vehicle and passenger safety.
Beyond hardware, NVIDIA has cultivated strong partnerships across the autonomous driving ecosystem. Over 225 development partners, including companies like Zoox, OptimusRide, and Nutonomy, are using the DrivePX platform. In 2018, NVIDIA partnered with DHL and ZF to build an unmanned logistics fleet, leveraging ZF’s ProAI system and NVIDIA’s DrivePX platform.
Additionally, NVIDIA launched the DriveIX SDK, aimed at helping developers create advanced features such as driver/passenger face recognition, distraction detection, and interior equipment adjustment. This software tool was expected to be released in late 2017 and would complement other critical functions needed for fully autonomous vehicles.
Despite these advancements, achieving SAE Level 5 autonomy remains a long-term goal. Current systems only reach Level 2 or 3, and the technologies required for full automation—such as 5G communication and infrastructure integration—are still evolving. As a result, there is still much work to be done before fully autonomous vehicles become widespread.
Unmanned vehicles are likely to first see mass adoption in public transportation services like taxis and buses. These environments offer controlled settings for testing and deployment, reducing risks and management costs. Public transport providers stand to benefit from lower labor costs, further incentivizing the shift toward autonomous solutions.
Ecosystem collaboration will play a crucial role in the success of autonomous driving. NVIDIA has built a strong network by working closely with automotive manufacturers, researchers, and developers. Through open innovation and strategic partnerships, NVIDIA has positioned itself as a key player in the automotive tech space, setting the stage for future advancements in self-driving technology.
I-type Inductance Core,I Inductor Model,Ring I-type Inductance,Design Of I-Shaped Inductor
Xuzhou Jiuli Electronics Co., Ltd , https://www.xzjiulielectronic.com