We all must miss playing in the water on hot summer days during the cold winter months.

I believe that for many friends who like to swim, the underwater world is much more exciting than the sky, we have seen the aerial flight omnipotent aerial photography, but how to shoot in the unfathomable bottom of the sea?

Marseille, France-based startup Notilo +, which is also a member of the NVIDIA Inception program, is using iBubble - the world's first autonomous underwater drone - to capitalize on this opportunity.

This underwater aerial drone called iBubble has a cute and nerdy design with powerful dive shooting performance.

The new wave of automated machines

So far, underwater drones have been expensive and require professionals to operate them, and the target audience is very specialized. But the iBubble is powered by a 7.5w NVIDIA Jetson TX2 supercomputer that serves divers, oceanographers, underwater maintenance personnel, boat owners, and more.

Notilo +'s 15-person team shares their love for underwater adventures, so they know how complicated underwater filming can be. They designed the iBubble to allow divers to spend more time exploring and less time behind the camera. Thanks to a combination of data-heavy training and processing power, underwater unmanned submersibles can make their own decisions and follow a diver's route better than a human with a remote control.

Divers can choose to photograph their surroundings, or capture themselves as they explore from a different location. They can also communicate with the drone via a remote device that allows them to call the iBubble and change the focus of the shot.

The important role of Jetson

The technologies that allow most connected devices to flourish, such as Wi-Fi, GPS and Bluetooth, don't work underwater. Most companies try to compensate for this with expensive acoustic technology.

But iBubble relies on cost-effective sensors coupled with powerful machine learning algorithms. It combines acoustic data with computer vision to create a fast, reliable tracking system capable of handling the most difficult environments.

The Jetson TX2 makes it possible. With advances in machine learning techniques, embedded AI computers can implement real-time detection of convolutional neural networks. It uses LSTM networks to predict the sound signal, which greatly reduces the delay in signal processing.

The Jetson TX2's advanced multimedia processing capabilities support HD streaming. The powerful Jetson module can perform visual tracking as well as handle traditional reflected acoustic signal processing. Conventional signal processing is critical as it allows iBubble to not only detect and avoid obstacles (such as reef structures), but also to find and inspect areas on the vessel that need repair.

"You have to work with Nvidia when it comes to building edge computing solutions." Nicolas Gambini, CEO of Notilo+, said, "It's not just the benefits in terms of computing power, but also the resources NVIDIA provides to make it easier for startups to get started, which is unparalleled."

Recommended>>
1、Getting Started with Regular Expressions
2、Performing data classification by warping space a novel neural network architecture based on vector fields
3、A new type of worm known as the gold miner
4、Excel calculation formula book
5、The Unicorn of online taxi platforms

已发送

发送中