October 12, 2017

Hunting Asteroids with Data Science

Today, a small asteroid, 2012 TC4, will pass very close to Earth. It was originally discovered in October 2012 as it sped by our planet at approximately one-fourth the distance from Earth to the Moon. 2012 TC4 is one of more than 16,000 near-Earth asteroids (NEAs) that regularly pass by our neighborhood in space. A NEA is simply an asteroid with an orbit that brings it close to Earth. While some of these objects have the potential to collide with our planet, here at Planetary Resources we focus our time determining which ones are the most valuable for mining purposes.

TC4 Orbit. Photo Credit: NASA JPL

We have previously written about the initial steps in our prospecting program: evaluating asteroid characteristics such as size, spin, gravity, and composition. These factors help us decide which asteroids might be potentially valuable targets for mining water which we plan to refine into rocket propellant. These findings represent the best “first pass” we can make using ground-based telescopes and existing data sets.

Shape model of an asteroid created using ground-based radar observations where red indicates areas of higher gravity. Photo Credit: NASA JPL

The second phase of the evaluation involves our data science team evaluating the target to determine viability as a mine. Our team is comprised of data scientists, mission designers, and business analysts, all working together to estimate the value of a fully operational mine located at each potential target. We use the asteroid’s orbit to model all the possible outbound and return trajectories for a huge spacecraft we’ve nicknamed the  “ice barge”. The ice barge is a massive in-space vehicle that will move mined ice from the asteroid back to the propellant refinery. The refinery will be an orbital facility gravitationally-located near the actual point of need, potentially near the Moon, that converts the water-ice into our primary products: liquid oxygen and liquid hydrogen, the components of high-efficiency rocket fuel. Our team’s evaluation uses all possible trajectories to calculate the most profitable set of fuel deliveries from a mine using an algorithm similar to what Google Maps uses to minimize travel time between two locations. However, in our case, we are evaluating thousands of potential trajectories and attempting to minimize the time and energy needed to travel between the target and the refinery. This is a powerful application of data science that ties the economic metrics of a potential asteroid mine to an orbital supply chain algorithm.  

Reconnaissance is the first step in the exploration process, which typically involves an iterative series of detailed investigations, seeking to characterize the target prior to developing a mine. Now that we have selected numerous targets as a result of the prospecting program, we can move to the exploration phase. Our Exploration Program will begin in 2020 with the deployment of multiple spacecraft via a single rocket to visit our targeted asteroids. Our Arkyd-301 spacecraft will determine the ore grade and tonnage of the asteroid, helping us determine the value of the water resources. The spacecraft will also conduct a detailed environmental assessment to inform the future mine design.

The Arkyd-301 Exploration Spacecraft

In order to determine the potential value of the resources, we again turn to our data science team. Traditional mines here on Earth use detailed geostatistical models to estimate how much ore is buried underground (the ore body), where its boundaries are, and what the possibilities for extraction might be. In our case, the actual ore body is the entire asteroid and so our goal is to determine how much water likely exists in the asteroid in order to assign a value. Using data we collect showing indications of water from hydrated minerals on the asteroid’s surface, we can use models similar to those developed by the terrestrial mining industry to estimate the total amount of water within the asteroid. As we learn more about the asteroid, our initial estimates of the total water can be refined, increasing our confidence in the value of the deposit.  

Establishing a viable asteroid mine is similar to a traditional mine here on Earth in that it requires similar levels of exploration and due diligence. Today our data science team continues hunting for the best mining targets as we prepare for the first commercial deep-space exploration program in history. These are truly exciting times to be alive, and asteroid mining is one of the most critical steps we can take to becoming a true spacefaring civilization.


John Shriver, Data Scientist

James Orsulak, Director of Business Development and Amateur Asteroid Cartographer