10.7 C
Los Angeles
Friday, March 24, 2023

Nvidia GTC Highlights The Physical Side Of AI

- Advertisement -

- Advertisement -

ChatGPT launched a tide wave of interest in AI. For many consumers, AI is finally living up to long overdue expectations. The accomplishments of ChatGPT in a short period of time are phenomenal. But what is yet to come when AI is combined with robotics will change everything.

- Advertisement -

I have been promoting the advances in robotics for several years. I even called 2022 the year of robotics, partially because of the growing need to overcome shortages in labor and to handle tasks beyond the physical or mental capability of humans, and partially because of the continued advances that AI, accelerated processing, semiconductor, sensors, wireless connectivity, and software technologies are enabling to develop advanced, autonomous machines. Robots are no longer just for the manufacturing floor. They are hazardous material handlers, janitors, personal assistants, food preparers, food deliverers, security guards, and even surgeons that are increasingly autonomous. Essentially, they are AI in the physical world. As a result, robot competitions are heating up from middle schools to Las Vegas.

As seen at CES, robotics technology is advancing rapidly with advances in technology. My favorite examples were the multi-configurable Yarbo outdoor robot and the John Deere See & Spray. Yarbo can be a mower, a leaf blower, or a snow blower. If it could dispose of animal excrement and the annoying neighbor, it would be perfect yard tool. On the other end of the spectrum was the John Deere See & Spray Ultimate, a tractor with up to a 120-foot (36.6m) reach that uses AI/ML to detect weeds smaller than the size of a smart phone camera and spray herbicide accordingly. John Deere also offers self-drive tractors.

There is a great deal of new technology going into these platforms and some of the latest of these technologies will be on display next week at another key AI technical conference – The GPU Technology Conference (GTC) – hosted by Nvidia. GTC started out as a gaming conference but has morphed into one of the premier AI conferences, of which a key component is robotics/autonomous machines.

One of the reasons for this transformation of GTC is Nvidia’s continued investment in both AI and robotics technologies, as well as in the ecosystem. At CES, Nvidia announced a new tool called Isaac Orbit for simulating environments and benchmarks for robot learning and motion planning. Nvidia also announced enhancements to Isaac SIM for reinforcement learning, collaborative programming, and realistic human simulation. Nvidia’s commitment to the segment is demonstrated in new announcements to the platforms and tools every few months. As with previous GTCs, we expect new platform and tool announcements next week.

In addition, GTC is offering a track just for robotic developers that will feature some of the applications using or being developed on the Nvidia platforms, current and future software tools, ecosystem partners and solutions, and new Isaac ROS (robot operating system) features for the open source community. These sessions include topics like “Training Highly Dynamic Robots for Complex Tasks in Industrial Applications” and “Design a Complex Robot Architecture on NVIDIA Isaac ROS.” Tracks will also include tutorials on how to use the various tools and the latest tool features for the Tao developer framework, Deepstream for AI vision applications, and the Isaac SIM for development, training, and testing. However, with all of Nvidia’s hardware, software, and tools built on the same AI technologies, other topics may be of interest to robotic developers, such as the Metropolis framework for AI video analytics and it’s Omniverse collaboration environment. Even the Nvidia GPU roadmap points the way to future Jetson platforms.

Unfortunately, this GTC will still be virtual, so we won’t get any of the up-close robot demos, but the event is free with open access to replay the on-demand sessions. The most important part of GTC for robotics is that the announcements at GTC set the stage for other announcements in robotics through the year.

Source link