Nvidia Aims To Power Flying Vehicles with Jetson TX1 Board
The Jetsons, that space-age cartoon family, have finally arrived, complete with flying vehicles. Well, actually it's Nvidia’s Jetson TX1 module (pictured), designed to allow drones to fly autonomously without the need for human operators, that has arrived. The credit-card-size module, which debuted yesterday, aims to bring the power of machine learning to UAVs (unmanned aerial vehicles), robots, and other smart devices.
The potential for the platform is considerable. If Nvidia’s claims are accurate, the Jetson TX1 will allow autonomous vehicles to navigate through forests for search and rescue missions, scan crowds for suspicious activity, and tailor their behaviors to users' preferences.
Big Step in Autonomous Capabilities
The company said that the Jetson TX1 is designed to process deep neural networks, allowing it to recognize objects and interpret information. The embedded computer incorporates capabilities such as machine learning, computer vision, and navigation into a single system, greatly expanding the ability of machines to operate on their own and adapt to their surroundings.
As a result, the platform is capable of performing complex tasks such as recognizing images, processing conversational speech, or analyzing a room full of furniture and finding a path to navigate across it. Nvidia described the Jetson TX1’s machine learning abilities as “a groundbreaking technology that will give autonomous devices a giant leap in capability.”
Despite the module’s relatively small size, the company said it can perform a full teraflop of calculations, comparable to the computing power available to supercomputers at the turn of the century, while drawing very little power. The Jetson’s energy efficiency should help open up a number of applications for any UAV it powers.
Taking AI Out of the Lab
The Jetson TX1’s combination of lower power needs and machine learning abilities will allow developers to essentially put AIs into the field. “It has uses that range from artificial intelligence-assisted robots, to advanced systems in automobiles, and to Internet of Things-connected intelligent machines,” Ovum principal analyst Michael Azoff said in a statement. “The ecosystem around Jetson will accelerate the transfer of AI from lab to real-world machines."
Another potential advantage of the Jetson is that it's easier to develop for than most embedded boards. Because the module’s GPU supports Nvidia's CUDA interface, developers will be able to use its GPU to accelerate applications without having to learn more complicated GPU programming languages. Developing for the Jetson is more akin to developing software for a PC, according to Jeff Bier, president of Berkeley Design Technology.
The platform will come with a 256-core Maxwell architecture-based GPU, along with a 64-bit ARM A57 CPU. It will also support 4K video encode and decode and 1400 megapixels/second video recording, along with 4 GB of memory, 16 GB of storage and support for the Linux OS.
The Jetson TX1’s Developer Kit will be available for preorder starting tomorrow for developers in the U.S., with availability in other regions coming in the next few weeks. The module itself, meanwhile, will be available worldwide in early 2016 at a suggested price of $299.
Read more on: Nvidia
, Artificial Intelligence
, Machine Learning
, Mobile Apps
, Next Big Thing
, Tech News
Posted: 2015-11-19 @ 11:16pm PT
So this module can upgrade a multicopter to a drone.