How AI Will Help the U.S. to Mars and Beyond

Technology has evolved rapidly since Neil Armstrong and Buzz Aldrin first walked on the moon in 1969. Even the technology that landed NASA's Pathfinder mission on Mars in 1997 wouldn't stand up to the smartphone computers we now carry in our pockets. 

NASA's Artemis program plans to send more astronauts to the moon in 2024 and eventually progress to Mars. Engineers have spent years developing and refining technology to ensure that these missions safely and efficiently gather the information we need for further exploration.

The future of space exploration will heavily rely on software systems, artificial intelligence and machine learning to predict conditions, object movements and make the technology we've spent so many years developing gather more information in less time. 

Let's take a look at some of the ways AI and ML will play a role in the future of space exploration. More specifically, looking at how AI will enable humans to establish permanent colonies on Mars and beyond.

Space Weather Prediction

Mars is a very hazardous environment. It has no magnetic field, which means there's very little protection from solar flares and cosmic rays. This also means that it doesn't retain heat energy like Earth, causing extreme temperature changes from day to night. According to NASA, the average temperature on Mars is about minus 80 degrees Fahrenheit. In the wintertime, the poles can drop as low as minus 195 degrees Fahrenheit, while, in the summer, it can reach up to 70 degrees near the equator. 

Such drastic temperatures demand an accurate weather model that can alert and prevent the crew from adverse exposure. Today it's relatively easy for anyone to predict weather patterns on Earth because we have centuries of meteorological experience to reference. We don't have that luxury at Mars. For several years, orbiting Martian probes and rovers have been collecting vast amounts of weather data. Still, there's no way for a human to analyze the data fast enough to understand how the weather patterns change confidently. That's why we need AI.

Similar to how we have a weather app on our phones with predictions about how the day will unfold, AI and ML programs can analyze the weather data and make accurate predictions about how the weather on Mars changes and where and when it will be safe for humans.

Launch Window Prediction 

There is a need for more accurate launch windows on the same token as predicting weather patterns to create safer space exploration. Even today, this is an issue on Earth with a relatively stable and forgiving climate, but still, we witness a high frequency of scrubbed or delayed launches. To mitigate the chance of leaving supplies or humans stranded on Mars, we need a highly robust and reliable system. It should determine, many weeks in advance when we can or cannot launch on demand. Drones, rovers and satellites are already making consistent trips to space, and humans may be soon to follow as we learn more about where to travel and land. 

Launch Window Prediction is a developing technology that uses AI in conjunction with a weather prediction model to safely determine critical launch conditions and give launch commands like GO/NOGO.

Years in the future, if rockets are traveling between the Earth and Mars, the same technology is even more important for return trips from the red planet. This is why it's so critical that AI and ML are busy analyzing weather data on and around Mars to accurately predict when it's safe for a rocket to be there. 

Systems Control Automation

One of the driving factors to further space exploration is efficient communication with the rovers in space. Right now, the NASA Jet Propulsion Laboratory is doing a lot of computer vision and autonomous driving with the Mars 2020 Perseverance Rover Mission. There's roughly a 20-minute communication time delay between the rover and NASA scientists, which slows down their research. 

If the best a robot can do is 20 minutes, imagine the communication delay with people on Mars instead of a robot. Flight surgeons, mission directors and the entire support team will not have real-time instant communications with Martian astronauts like we do today with the Space Station. 

To streamline research, all communication between Martian astronauts and NASA would require some kind of automation. The chance of failure amid uncertainty is so high; we can't risk letting a single person manage and maintain the crew's livelihood. Eventually, we can bring flight surgeons, mission directors and the entire support team to Mars, where they work with field researchers and astronauts. However, we cannot afford to bring 5 to 10 times the number of people that style of familiar collaboration would require in the near future. Instead, we must look to AI companions and support robots.

An example of this AI robot/human relationship can be found in “2001: A Space Odyssey” film. HAL 9000 (perhaps SAL 9000, the friendly version) is a robot-human companion that monitors the crew and colony's health and activities. It manages food production, task and repair management, and science goals and directives. 

Innovative technology is what landed us on the moon over 50 years ago. The same mindset will get us to Mars. Still, it requires a shift in how we employ technology to help us make missions safer and deliver faster results, focusing on developing programs to inform launches first and then hardware to house them.

Modern AI and ML technologies push the limits of what was previously thought possible for space exploration. These are just a few examples of concepts and programs in development today. It's not so much about embracing the unknown as it is teaching ourselves about it before we conquer it. 

Michael Limotta is the co-founder and an AI Architect for Aerospace and Physics at Proximai.