How about helping NASA train an algorithm that identifies structures on Mars?

Although artificial intelligences have enormous potential to help scientists explore the universe, they need human help to be trained in the process. Thus, NASA invites the public to help in this mission. Through the AI4Mars project, participants can help an artificial intelligence algorithm recognize formations of scientific interest in the Red Planet photos taken by the rover Perseverance.

AI4Mars is, in fact, a project that is a continuation of another one launched last year, which resulted in the identification of formations in almost half a million images. To do this, participants used an online tool to highlight interesting formations, such as sand and rocks, analyzed by NASA scientists to determine possible dangers in the path of the rovers on Mars.

Project participants outline and identify different rocks and formations in the landscape of Mars (Image: Reproduction/NASA/JPL-Caltech)

The work resulted in the creation of the Soil Property and Object Classification (SPOC) algorithm, capable of identifying structures with 98% accuracy. As SPOC is still under development, scientists plan to use the images taken by the Perseverance rover to enhance it, expanding the types of tags that can be applied to Martian surface formations.

Want to catch up on the best tech news of the day? Access and subscribe to our new youtube channel, Canaltech News. Everyday a summary of the main news from the tech world for you!

For this, AI4Mars has identification tags for finer details, for participants to choose options such as floating rocks (“islands” of rocks) or nodules (structures formed by water and cemented minerals). Thus, in the future, the algorithm could help the rover to select information of interest from the pile of data sent to Earth – which could involve hundreds of images daily, which need to be analyzed in a few hours for scientists to devise a new set of instructions for the rover .

Annie Didier worked on the AI4Mars version aimed at the Perseverance rover and already suggests some purposes for the Mars data. “With this algorithm, the rover can automatically select scientific targets and go to them,” she proposed. Furthermore, the algorithm could store assorted images on board the rover, and then send back images of structures of interest to scientists.

Parts of the Perseverance rover visible alongside a highlighted area in the project (Image: Reproduction/NASA/JPL-Caltech)

Vivian Sun, a scientist at JPL, explains that it is impossible for scientists to analyze all the images accurately in such a short time, every day. “If an algorithm could tell us, ‘I think I saw rocks or nodules there,’ we would have great time savings and the scientific team could analyze those areas in more detail,” he said. Even so, it is worth remembering that even when it is improved, the algorithm does not replace the more complex analyzes carried out by human scientists.

To participate in AI4Mars, just click here.

Source: JPL

Did you like this article?

Subscribe your email on Canaltech to receive daily updates with the latest news from the world of technology.

About Raju Singh

Raju has an exquisite taste. For him, video games are more than entertainment and he likes to discuss forms and art.

Check Also

Genshin Impact Codes for Today, December 4th, 2021

codes Genshin Impact for today, December 4, 2021. Redeem right now, remember codes can expire …