NVIDIA Introduces Chip To Empower AI And Robotics In More Industries
The Jetson Xavier Computer Provides Same Processing Power As A $10,000 Workstation
NVIDIA’s game plan of applying its GPU in the Artificial Intelligence (AI) specimens has helped it gain mileage in the technology advancement journey. The company is doing pretty well in rapidly growing fields like data centres and self-driving cars. Its eyes are focused on incorporating AI technology in fields like agriculture, manufacturing, logistics and construction, all of which have required maximum attention from human engineers before.
Recently NVIDIA announced NVIDIA Isaac, which the company has elaborated that it will be the new platform that will act as the artificial intelligence backbone to a generation of autonomous machines. The heart of Isaac platform is NVIDIA’s new Jetson Xavier computer which is the world’s first computer designed specifically for robotics.
At Computex 2018, Jensen Huang who’s the company’s CEO said: “both Isaac and Jet Xavier are designed to capture the next stage of AI innovation as it moves from software running in the cloud to robots that navigate the real world.” He further added, “AI, in combination with sensors and actuators will be the brain of a new generation of autonomous machines. Someday there will be billions of intelligent machines in the manufacturing, home delivery, warehouse logistics and much more.”
Isaac Platform is a combination of software tools that will enable companies to incorporate AI systems so as to develop and train robots. At its heart is NVIDIA’s new Jetson Xavier which is equipped with more than nine billion transistors that deliver over 30 trillion operations per second. The worlds first robotic computer ironically consumes only 30 watts of power, a figure that is half of what the bulb in your room consumes. The company’s CEO, Huang, claims the computer provides the same processing power as a $10,000 workstation! The new server platform can replace 300 dual CUP server nodes in terms of the performance of a computer.
Jetson Xavier has six superior processors that include dual NVDLA deep learning accelerators, Volta Tenser GPU, an image processor, a video processor and an 8-core ARM64 CPU. It can process tons of algorithm in short time for works like odometry, localization and mapping and sensor processing, in a simultaneous sequence.
Together with access to Isaac platform, one Jetson Xavier will cost $1,299, something that makes it undoubtedly true that this hardware will offer a lot of power at an honest price. Sources close to the industry expect the new platform to be built using TSMC’s 12nm process and enter the market in early August. NVIDIA still has a lot to do in order to catch up with Intel, which still has the dominance in the gaming market.
The company is currently collaborating with a few OEM and ODM players on developing HGX-2 based systems for the world’s datacentres like Wistron, Asustek, Quanta, Foxconn and Wiwynn.