We dream of exploring strange new worlds and discovering new lifeforms — ideally without having to do all the work ourselves. NASA’s Perseverance Mars Rover is leading the way by giving us a taste of what can be achieved with AI in space.
With the help of its onboard AI, the Perseverance rover is engaged in a unique kind of treasure hunt across the Red Planet’s rocky landscape. But instead of gold and jewels, it’s looking for something arguably more valuable: minerals.
Perseverance, AI and Martian Minerals
The rover has been using artificial intelligence (AI) to search for minerals in Martian rocks for nearly three years. It is the first Mars mission ever to use AI for autonomous, real-time analysis of rock composition.
Behind this innovative use of AI is the Planetary Instrument for X-ray Lithochemistry (PIXL), a spectrometer developed by NASA’s Jet Propulsion Laboratory (JPL) in Southern California.
According to Abigail Allwood, JPL’s principal investigator for PIXL, this clever instrument uses the artificial intelligence aboard Perseverance to find interesting minerals in Martian rocks in real time.
“We’re using PIXL’s AI to focus on important science,” Allwood explains. “Without AI, you’d see a hint of something interesting in the data and then have to rescan the rock to study it further. This allows PIXL to draw a conclusion without humans examining the data.”
Another plus for Perseverance is that it can determine when it’s time to drill a core of rock and save it for future study on our home planet.
It’s all part of NASA’s ambitious Mars Sample Return campaign, a critical component of our efforts to understand the geologic history of Mars and, potentially, its capacity to host life.
“Advanced Sampling” and the Power of AI
Perseverance’s AI-driven approach to mineral exploration is known as “adaptive sampling.” Simply put, this means the rover’s AI scans a rock, identifies minerals of interest, and then autonomously decides where to focus its investigation.
All of this is done without the need for contact with mission control on Earth, saving time and resources.
Peter Lawson, who led the implementation of adaptive sampling at JPL before his retirement, puts it in perspective.
“The idea behind PIXL’s adaptive sampling is to help scientists find the needle in a haystack of data, freeing up time and energy to focus on other things,” Lawson explained. “Ultimately, it helps us collect the best science faster.”
Adaptive sampling helps scientists find the indispensable needle in the haystack of data, allowing them to focus their attention and energy elsewhere.
The result? High-quality Mars science, delivered at an accelerated pace.
AI helps Perseverance take the perfect photo
AI enhances PIXL’s functionality in two key ways. First, it ensures accurate positioning of the instrument once it is near a rock target.
PIXL’s spectrometer is mounted at the end of Perseverance’s robotic arm and attached to six small robotic legs, known as a hexapod.
This complex setup allows PIXL’s camera to continuously monitor the distance between the instrument and the rock target, enabling precise positioning.
The large temperature swings on Mars cause Perseverance’s arm to expand or contract very slightly, which could prevent PIXL from aiming properly.
To counter this, the hexapod autonomously adjusts the instrument, allowing it to reach the vicinity of the rock without making contact.
“We need micrometer-scale adjustments to get the accuracy we need,” Allwood explained. “PIXL gets so close to the rock that it makes an engineer’s hair stand on end.”
This synergy between AI and robotics is an example of the advanced technology at the heart of our Mars exploration missions. It highlights the precision and adaptability needed to navigate and study the Red Planet’s challenging environment.
Final step: creating a mineral map
Once PIXL is positioned, another advanced AI system takes center stage. PIXL meticulously scans a postage-stamp-sized chunk of rock, using an X-ray beam thousands of times to create a grid of microscopic dots. Each dot provides a wealth of information about the chemical makeup of the minerals within.
Understanding these minerals is crucial to answering deep questions about Mars. For example, scientists could look for carbonates in certain rocks, which could reveal how water once formed them.
Another possibility is that they might be looking for phosphates, potential food sources for ancient microbes on Mars, if such life ever existed.
It’s impossible to predict which of the hundreds of X-rays will reveal specific minerals. However, when PIXL identifies certain minerals, it can autonomously stop to collect additional data, a process known as a “long dwell.” As machine learning improves the system, the repertoire of minerals that PIXL can analyze with a long dwell continues to expand.
The Curiosity Rover is also involved
But don’t think Mars is the only place where artificial intelligence (AI) is in space. NASA’s Curiosity rover is already deploying it from over 2,300 miles away.
From autonomously lasering rocks based on shape and color to self-navigation, Curiosity is a pioneer in AI-powered space exploration.
Its successor, Perseverance, has raised the stakes with more advanced AI capabilities, but mission planning is still largely a human affair, with dozens of engineers and scientists involved in the planning process.
AI and the Future of Space Exploration
As AI makes big leaps forward in field tests on Mars, David Thompson, a software engineer at JPL, emphasizes that these advances are especially important for future missions that venture deeper into the solar system.
“PIXL is like a Swiss Army knife, adaptable based on what scientists are exploring at any given time,” Thompson explains. “Mars provides an excellent testing ground for AI, because our daily interactions allow us to continually refine and improve the system.”
Future missions that go deeper into the solar system will face longer communication delays compared to current Mars missions.
Therefore, it is crucial to increase the autonomy of these missions, so that they can conduct independent research and science for the common good of humanity.
As communication windows for more distant missions shrink, the importance of autonomous AI will only increase, paving the way for the next generation of smart spacecraft.
As we continue to push the boundaries of space exploration, it’s becoming increasingly clear that our dreams of exploring the universe may not be so far away after all.
Thanks to the continued efforts of dedicated scientists like the team at NASA’s JPL and the power of AI, we are getting closer to unlocking the mysteries of our solar system. After all, who knows what mineral treasures lie in the rocks of distant worlds?
—–
Like what you read? Subscribe to our newsletter for engaging articles, exclusive content and the latest updates.
Check us out on EarthSnap, a free app from Eric Ralls and Earth.com.
—–
