Home > Science and technology > Aerospace
Reading time estimate: 2 min

Train NASA's astronauts to persevere

BingMag.com Train NASA's astronauts to persevere

NASA in a new project has asked all people to participate in the training of persistent astronauts to better identify the surface features of Mars.

However, this is just a robot, and sometimes Human understanding can help a robot become intelligent. If you, too, want to help a persistent astronaut, NASA will ask those interested in assisting with persistent machine learning algorithms. Mark them. This is something most people can do directly, but it can be difficult for a robot. Which began last year for the Curiosity astronaut. Curiosity arrived on Mars in 2012 and has been making history ever since. NASA used the experience of the Curiosity astronaut as a starting point when designing perseverance.

The new astronaut has 23 cameras that capture large amounts of image data from Mars, but for further interpretation of these images, the robot It relies on human operators. This astronaut has artificial intelligence to cross the surface obstacles of Mars well, and in NASA's new project with the help of people, its intelligence will be even better.

BingMag.com Train NASA's astronauts to persevere

Selfie of a persistent astronaut with a genius Mars helicopter
Credit: NASA/JPL-Caltech/MSSS/Sen Doran

For example, the NavCam asks the user for sand, integrated soil (which cycles Have good traction), identify bedrock and large rocks. Examples of all of these structures have been put in place, so it will be very quick to get started.

With all of this labeled data, NASA can better train neural networks to detect the surface features of Mars. Eventually, an astronaut may be able to spin around and collect samples without waiting for instructions from the mission control team and receiving detailed instructions for the slightest movement. Artificial intelligence training also helps to better identify the most important geological features. Saves man from blindly exploring several gigabytes of data.

The result of the AI4Mars Curiosity project is an algorithm called Soil and Object Classification (SPOC). The algorithm is still in active development, but NASA says it can now identify Martian features with 98% accuracy. And they are the bedrock texture, improving the SPOC algorithm again. In some images, almost all objects are pre-tagged, but others may be incomplete.

The Curiosity AI project recorded about half a million tagged images. Although tagging 20,000 images of perseverance is also a good achievement for the mission team, this amount is likely to be much higher.


Source: Extreame Tech

READ NEXT IN: science and technology / aerospace

BingMag.com According to NASA, the recent incident did not damage the James Webb Space Telescope aerospace

NASA has delayed the launch date by a few days to further investigate, citing an incident with the James Webb Space Telescope. "Engineering teams have conducted additional tests to confirm that NASA's

BingMag.com Scientists use Kepler telescope data to discover new multipolar systems aerospace

Astronomers using photometric data from NASA's Kepler telescope mission 2 have identified 747 potential planets and 57 multi-planetary systems with planet-like planets. Of these, 366 were not even pre

BingMag.com The Insight probe detected different structures beneath the surface of Mars aerospace

A closer look at the Earth's surface can tell us a lot about its history and geological structure, and the same is true of other planets like Mars. Now NASA's InSight probe on Mars has opened our eyes

BingMag.com Numerous false fossils on Mars may challenge life search aerospace

According to researchers, Mars may be covered with dozens of different non-living fossil fossils, which make the search for life on the Red Planet difficult. NASA's Perseverance spacecraft lands on Ma