NASA and Microsoft’s AI detection system, astronauts only need a few minutes in space

In many space-related movies, if an astronaut encounters danger, there is a high probability that the spacesuit will be damaged. For example, in "Interstellar", Cooper's helmet was cracked by Dr. Mann.

▲ Screenshot of the movie "Interstellar", the picture comes from: bilibili

When playing games, we often see the mechanism of setting status values ​​for items. We can intuitively see the available values ​​of these equipment, such as the content in the oxygen cylinder, the remaining attack power of weapons, and so on. However, in the mysterious and dangerous space, astronauts who wear heavy space suits for out-of-vehicle activities cannot accurately and timely grasp the situation of space suits as in the game.

Spacesuits are prone to hard-to-find wear and tear during extravehicular missions, which is a big safety hazard for astronauts. If it is not detected in time, it may lead to catastrophe.

▲ Picture from: Microsoft

In 2007, American "Endeavour" astronaut Rick Mastracchio in the extravehicular spacewalk, through the spacesuit helmet camera scanning inspection, found that both gloves were damaged. To be on the safe side, ground controllers ordered him to return to the space station urgently.

It can be seen that the safety of space suits and astronauts should be checked in time to avoid risks, especially the gloves that are prone to problems in space suits. NASA (National Aeronautics and Space Administration) has a set of glove safety inspection procedures, but because of the need to take pictures first, and then transmit data between the ground and space, each inspection takes a long time.

▲ Picture from: SpaceRef

At present, the scope of astronauts' activities in space is mainly around the space station and low Earth orbit, but human ambitions are far from this. In the future, we will go to the moon, Mars and even further deep space.

The communication between Mars and Earth has a time difference of about forty minutes. NASA's glove detection method, which has been used for 20 years, occupies too much communication transmission, which is obviously not enough to support exploration into the depths of space.

▲ Picture from: Microsoft

So a Microsoft team, working with NASA scientists and Hewlett Packard Enterprise (HPE) engineers, developed a system that leverages AI (artificial intelligence) to perform critical inspections of spacesuit gloves.

The developers used raw images taken in each of the EVAs (extravehicular missions) that previously helped build the International Space Station as the initial training dataset, ranging from 2,000 to 3,500 images. In addition, the dataset was augmented by photographing and inspecting damaged gloves to flag specific types of wear with gloves that were worn during spacewalks and training on the ground.

▲ Picture from: Microsoft

The AI ​​and ML automated glove inspection process trained on the dataset takes pictures of the astronauts' gloves when they take off their spacesuits after a spacewalk, and these pictures are immediately sent to the HPE Spaceborne Computer-2 ( Onboard Computer-2).

Spaceborne Computer-2, an artificial intelligence platform developed by Microsoft, HPE and NASA, can perform more than 2 trillion calculations per second. With it on the International Space Station, astronauts can increase the speed of processing data to just a few minutes from a distance of 227 miles (about 365 kilometers) from Earth.

▲ Picture from: Microsoft

Once the computer receives the photo, it quickly looks for signs of damage. If any damage is detected, a message is immediately sent to Earth highlighting areas for further review by NASA engineers.

There is no doubt that this new method of glove detection will greatly benefit astronauts' space activities. However, the technology, which is still in the development stage, will analyze the gloves but not make corresponding safety decisions. Therefore, it is necessary to continue testing to verify its reliability.

▲ Picture from: Microsoft

Although it is still necessary to continue to use the old method for inspection at this stage, the new technology represents new hope, not only for gloves, but may be extended to the inspection of other components in the future. Let human beings step into space, go faster and farther.

#Welcome to pay attention to the official WeChat account of Aifaner: Aifaner (WeChat: ifanr), more exciting content will be brought to you as soon as possible.

Love Faner | Original link · View comments · Sina Weibo