With Canada, the UAE lunar rover will test the first artificial intelligence on the moon

 By Elizabeth Howell

January 28, 2023

One enormous advance in artificial intelligence is set to occur on the moon (AI).

On a United Arab Emirates rover that launched with SpaceX on December 11, a Canadian machine learning system will travel there.

This spring, the Japanese ispace lander carrying the Rashid rover, which will explore the lunar surface in search of minerals and other interesting objects, is anticipated to land. In a significant first for AI, Canada's system will guide the rover's decision-making: According to industry executives, no AI has ever been beyond low Earth orbit.

According to Ewan Reid, CEO of Mission Control Space Services (MCSS), if this is successful, the technology will be crucial for NASA's moon mission. Reid predicted that "AI will be a major enabler for decision-making on board spacecraft." Making Earth observation more effective is part of that endeavour, in addition to NASA's intentions to use the Artemis missions to look for water on the moon. And MCSS, a business with only 40 employees as of now, wants to be in charge.

Most people think of flamboyant initiatives like Canadarm, the robotic arm series that has supported the shuttle, the International Space Base, and will soon NASA's Gateway lunar station when they think of Canadian technology in space. Other tech areas where Canada excels include space medicine and even rocketry, according to more observant observers.

However, MCSS is an example of what many tiny Canadian space firms do, which is to offer critical software or components that operate in the background and support missions all over the world. According to Reid, MCSS clients frequently desire public emphasis on their own purpose, thus the organisation doesn't market all of its activities.

This [AI] demonstration with the Canadian Space Agency is advantageous since it is mostly in the public domain and we are permitted to discuss it, according to Reid.

Rashid will likely operate for around one lunar day (29 days on Earth) if all goes according to plan. Although it is not anticipated to endure the full night, MCSS is perfectly content with that as a demonstration mission. It intends to make the most of a few priceless days spent on the moon.

The Japanese lander, which will manage communications with Earth, will send the Rashid rover's navigational pictures to MCSS. According to Reid, "every single pixel in the image [will be] categorised as a specific terrain type" using the Canadian company's technology.

"That output will then be delivered to the ground and will be used by scientists and engineers in our Ottawa headquarters, as well as at other Canadian universities," he added. Once engineers are convinced the AI can discriminate between various minerals and between crucial mission objects like rocks and craters, future missions will take the human "out of the loop."

Since AI would only communicate the data, pictures, and videos that scientists need on the moon, Reid added, it might save a significant amount of the finite satellite bandwidth. The technology can be used in many solar system applications, such as removing clouds from planetary photos or identifying "dark" ships on Earth that are attempting to sail without being registered.

Reid stated, "We see a huge opportunity for putting AI on the edge, in space. MCSS will be able "to help other firms and organisations as they try to employ AI in their missions in the future," if all goes according to plan, thanks to the lunar demonstration.

Under the terms of a Creative Commons licence, this article has been taken from Space.com. Go here to read the original article.