AI system checks astronaut's gloves for damage in real time

Microsoft and Hewlett Packard Enterprise (HSE) are working with NASA scientists to develop an AI system to test astronauts' gloves . Because the outer space environment is a very harsh environment, equipment failure is prone to occur and is catastrophic.

Gloves are often prone to wear and tear as they are used for everything, including repairing equipment and installing new equipment.

Currently, astronauts will send back an image of their glove back to Earth so that NASA analysts can conduct a manual check.

Tom Keane, Corporate Vice President for Mission Engineering at Microsoft, in a blog post said: "The process is complete with the ISS low-orbit distance of about 250 miles from Earth, but everything is fine. It will be different when NASA once again sends a man to the moon or to Mars - 140 million miles from Earth."

Description of the image returned by the analysis system from NASA.

Harnessing the power of HPE's Spaceborne Computer-2 , teams from three companies are developing an AI system that can quickly detect the slightest signs of wear and tear on astronauts' gloves that can affect astronauts' gloves. affect their safety.

The astronaut gloves are built to be sturdy and come in five layers. The outer layer has a rubber coating for grip and acts as the first layer of protection. Next is the layer of Vectran (a cut resistant material). The last three layers maintain pressure and protect from the extreme temperatures of space.

However, the outer space environment is too harsh and can overcome these protective layers. Gloves of astronauts face more dangers than what happens on the ground.

For example, micromaterials create many sharp edges on handrails and other components. When it comes to locations like the moon and Mars, the lack of natural erosion means the rock particles become more like glass than simply sand.

Gloves from space are subject to more damaging impacts than they are from the ground. (Photo: NBC News).

To create the glove analyzer, the project team first started with images of new, undamaged gloves and those featuring wear and tear from spacewalking and spacewalking. shallow. NASA engineers went through the images and tagged specific clothing types through Azure Cognitive Services Custom Visions.

The cloud-based AI system was trained using data, and the results were comparable to NASA's own actual damage reports. The tool generates a specific number of damage probabilities for the area of ​​each glove.

In space, an image of the astronauts' gloves will be taken as they remove their equipment in the capsule. These images will then be analyzed locally using HPE's Spaceborne Computer-2 for signs of damage, and if any, a notification will be sent to Earth with areas marked for engineers. NASA considers further.

How AI scans analyze astronaut gloves.

"What we've proven is that we can do AI and edge processing on the ISS and glove analysis in real time ," said Ryan Campbell, senior software engineer at Microsoft Azure Space . Because we can be with the astronaut while we're processing, we can run our tests faster than images can be sent to the ground.'

The project is a great example of the power of AI combined with edge computing, in areas of limited connectivity like space.

In the future, the project could expand to early detection of damage to other areas such as the installation of hatches before they become a serious problem. Microsoft even envisions that a device like the HoloLens 2 or a successor could be used to allow astronauts to visually scan for damage in real time.

Jennifer Ott, Data and AI Specialist at Microsoft concludes: 'Bringing cloud computing power to the top through projects like this helps us explore and prepare for what we can. do next in a safer way such as we expect future long-range human spaceflight and as we begin to push that edge further out '.