The hottest NVIDIA relies on two platforms to take

NVIDIA relies on two platforms to take the lead in the field of autonomous driving

NVIDIA relies on two platforms to take the lead in the field of autonomous driving

China Construction machinery information

according to foreign media the street, NVIDIA is one of the first companies to attend the 2018 autonomous vehicle conference held in Novi, Michigan, USA, This once again confirms the company's leading position in the field of autonomous driving

this innovative technology company has been on the market frequently recently, including the company's recently announced Turing GPU architecture, and its CEO Huang Renxun delivered a keynote speech in Cologne, Germany. The company released its quarterly financial report on August 16

although there were some fluctuations in the cryptocurrency business of the company in this quarter, the rest of the performance was very strong

data center, game and automobile businesses have shown impressive attraction, and show no signs of slowing down

specifically speaking, the company's quarterly revenue of this business increased by 13% to a record $161million. The sector grew by 11% in a row, achieving the first double-digit growth in three quarters

drive Pegasus platform

perhaps more exciting is that NVIDIA began to launch its drive Pegasus platform, on which Daimler and Bosch chose to establish their own self driving taxi service

drive Pegasus platform will support level 4 and level 5 autopilot capabilities. In fact, its self driving and "mobile as a service" platform are one step closer to implementation

Mercedes Benz has used NVIDIA to power its impressive mbux AI infotainment system to change the angle. The company will use drive Pegasus to support its self driving fleet

drive constellation platform

drive constellation, another product of NVIDIA, should be able to safely promote its automatic driving into reality. Drive constellation is a huge project that will allow developers to test their programs in the virtual world. Simply put, drive constellation will allow you to synthesize test environments

this means that, like most things in life, the quality of autonomous driving tests is more important than the quantity. There is no advantage in testing a system repeatedly on a straight and open channel. On the contrary, testing in a more realistic environment can bring more specific improvements in autonomous driving ability

although the more realistic test ring and the main electrostatic parts and sliding parts in the CFRT market are preferred, it needs to pay a price: time and safety. Simply put, it takes a lot of time to record mileage in the real world. While doing so, drivers are also facing the risk of potential accidents

however, in the virtual world, due to scalability, the learning time is greatly accelerated, and safety is no longer a problem. The specimen between the two fixtures will get varying degrees of tensile deformation, because the real driver has no risk. This may make some people doubt whether the virtual test can solve the problem of improving automatic driving

real world testing is still necessary. Synthetic testing is not a successful one-stop shop, but it will play a great role in accelerating technological development

companies such as waymo, a autonomous vehicle company under alphabet, have test vehicles that can drive tens of thousands of miles a day in the real world and 10million miles a day in the virtual environment

although waymo does not use drive constellation, it is clear that companies and mobile service providers using drive constellation can benefit more from it

assemble the nose in the following order. Considering that NVIDIA is cooperating with more than 370 automotive partners (not all partners will use drive constellation), it is obvious that when it is widely used, there will be more interested partners