NVIDIA Adds Next Layer Of AI Automation Development Platforms With Isaac

By Christopher Wilder - June 6, 2018
Project Isaac playing golf.

This week at Computex in Tapei,  NVIDIA announced the availability of its Isaac platform. This AI-based development platform is a combination of hardware, software, and a virtual-world robot simulator designed to power autonomous machines within many industries: agriculture, construction, manufacturing robotics, and more. This announcement is evidence of NVIDIA’s commitment to enabling and empowering developers to design, develop, and deploy AI-based solutions.

This announcement brings immediate benefit to NVIDIA's customers and AI developers across a wide range of industries. As AI evolves, it promises unprecedented automation improvements and better overall business efficiency. The convergence between hardware, sensors, actuators, and the software that drives the decision making continues to be more and more integrated. Further, as more of these intelligent devices are deployed, companies will be required to evolve their digital transformation strategies to meet customer demands of interoperability, integration, and intelligence amongst their service providers. AI is indeed ushering in one of the most significant technology phase shifts of our time. The evolution of Jetson Xavier continues NVIDIA Isaac, derived from Jetson Xavier, is touted as the first computer built for robots. At its core, Jetson Xavier has more than 9-billion transistors, over 30 TOPS (trillions of operations per second), a Volta Tensor Core GPU, an eight-core ARM64 CPU, dual NVDLA deep learning accelerators, image and vision processors, and more. Altogether, these features make it possible for robots and other devices to process dozens of algorithms concurrently to receive feedback, process information, and make decisions in real-time. Speaking as a traditional networking and telecoms guy, I continue to be amazed at how NVIDIA packs so much efficiency and power into a single computer, using one-third the energy of a lightbulb. The general availability date for Isaac is August 2018; it will include (taken from NVIDIA's release):
  • Isaac SDK – a collection of APIs and tools to develop robotics algorithm software and runtime framework with fully accelerated libraries.
  • Isaac IMX – Isaac Intelligent Machine Acceleration applications, a collection of NVIDIA-developed robotics algorithm software.
  • Isaac Sim – a highly realistic virtual simulation environment for developers to train autonomous machines and perform hardware-in-the-loop testing with Jetson Xavier.
Moving forward There is no doubt there will be an immediate benefit to NVIDIA and the AI developer community. AI applications and specific use cases are evolving at a rapid rate. In my opinion, NVIDIA is in an excellent position to provide the right balance of developer capabilities and will be the ones best-suited to grow with the needs of its customers and partners. The timing couldn't be better. NVIDIA is reaching for the thought leadership mantle within AI, and it is obvious the company is making significant strides. However, one area where NVIDIA could improve is demonstrating customer deployments, real-world use cases, and partnerships to help augment holes in NVIDIA's messaging—especially from a security and infrastructure perspective. NVIDIA is closing the capabilities gap within AI and is driving much of the thought leadership and innovation story, especially within the self-driving automobile, robotics, and smart manufacturing industries. However, several of NVIDIA's competitors are turning to partnerships to help close the gap. In many cases, these companies are doing an adequate job of demonstrating the business cases of privacy, security, and governance (especially within IoT) that NVIDIA has not adequately addressed. However, in its defense, NVIDIA is more about enablement than the applications/use cases that reside on its platforms. Its philosophy is let the developers create the use cases. Conclusion Jetson Xavier continues to evolve and become a more and more robust platform for developing robotic solutions. NVIDIA's Isaac should be a sound toolkit for designing and simulating the next-level of algorithms and functionality for industrial automation. Isaac is an exciting next step for robotics and AI. Although we are at the beginning of this exciting market, it sure is fun to watch NVIDIA continue to innovate and execute.
Website |  + posts