Should a Lunar Rover Grab a Good Sample – or Hold Out for a Better One?
Baylor researchers combine planetary geology with business analytics to help lunar missions make smarter sampling decisions
An illustration of the planned Endurance rover on the Moon. (Image credit: NASA)
Contact: Kelly Craine, 254-297-9065
Follow us: @BaylorUMedia on X and LinkedIn
Imagine sending a rover hundreds of miles across the Moon to collect lunar geological samples, with space to collect only one rock. Do you take the first promising specimen you find – or keep moving and risk ending the journey with something less valuable?
As scientists prepare for the next generation of Moon exploration, lunar rovers will play a critical role in discerning which samples are worth collecting and returning to Earth for study. These specimens hold clues to the geological makeup of the Moon and the formation of the solar system. But how do scientists program autonomous lunar rovers to make these high-stakes decisions? Baylor University researchers have come up with a solution using a surprising combination of business analytics, economics and planetary geology.
The study, “Optimal Rover Sampling Informed by Orbital Remote Sensing,” was conducted by Rory Jake Samuels, Ph.D., and David Dicks, Ph.D., of Baylor’s Hankamer School of Business and Peter B. James, Ph.D., associate professor of planetary geophysics in Baylor’s Department of Geosciences. The research was presented at the 57th Lunar and Planetary Science Conference in March 2026.
“We are at an exciting point in history where surface operations on the Moon are about to ramp up dramatically." - Peter B. James, Ph.D.
“We are at an exciting point in history where surface operations on the Moon are about to ramp up dramatically,” James said. “An increasing amount of exploration will be performed autonomously, and this work provides the first rigorous roadmap for autonomous sample collection.”
The interdisciplinary team developed an algorithm that instructs a lunar rover when to stop and collect a geological sample during a long journey across the Moon’s surface. In simulations, the researchers tasked the lunar rover with gathering samples with a high content of thorium, an element that can offer clues about the Moon’s composition and geologic history. The algorithm more than doubled the expected thorium content of the selected sample compared with a traditional decision-making strategy.
A 500-mile decision
Lunar rovers can gather valuable information as they travel, but they have limited capacity to store rock and soil samples. Choosing a sample too early could mean missing an extraordinary discovery farther down the route. Waiting too long, however, could leave the rover with a poor final option.
The Baylor researchers modeled an approximately 800-kilometer – or nearly 500-mile – route across the Moon’s South Pole-Aitken basin. The rover’s mission was to collect one sample containing the highest possible concentration of thorium.
To guide that decision, the researchers applied Optimal Stopping Theory, a mathematical framework for determining the best time to act when future opportunities are uncertain.
The algorithm used orbital maps of thorium-abundant locations to estimate what the rover could encounter later in its journey. At each location, the rover compared the thorium in the sample with the expected value of continuing to another sample. If the sample was valuable enough, the rover took it. If not, it moved on.
The standard for what qualifies as “valuable enough” changes throughout the trip. Early in the journey, when many opportunities remain, the rover can afford to be selective. As it nears the end, the qualification gradually drops.
When business analytics meets planetary geology
The decision facing a lunar rover resembles problems studied in economics, finance and operations research. In each case, a decision must be made with limited information and no certainty about what opportunities may come next.
“Business analytics is the science of learning from data, but for a purpose,” Samuels said. “The purpose is improving decisions. We have some collection of data, and it’s up to us to extract useful meaning from it. That meaning then flows into the decisions we eventually make.”
The connection had not been widely applied to rover sampling, illustrating how researchers in seemingly distant disciplines can recognize different versions of the same underlying problem.
“Those worlds had never collided – geosciences and economics – even though they were working on the same problems under different disguises or applications... The question was: Can we apply the same kind of mathematical solution to this completely different problem?” - Rory Jake Samuels, Ph.D.
“Those worlds had never collided – geosciences and economics – even though they were working on the same problems under different disguises or applications,” Samuels said. “We solve problems like this all the time in economics when we’re pricing options, for example. The question was: Can we apply the same kind of mathematical solution to this completely different problem?”
The collaboration also required the researchers to learn one another’s technical languages. Samuels and Dicks contributed expertise in statistics, probability and decision-making, while James provided knowledge of the Moon’s geology, orbital measurements and the practical constraints of planetary exploration.
“For the algorithm to work, it needs input from geologists about the distributions of various kinds of rocks, so there are opportunities for further collaboration with scientists,” James said. “There are some details about the lunar surface that we won’t know until we return, and that kind of new information will help refine the sampling strategy.”
More than double the expected payoff
The Baylor model improved the expected thorium content of the selected sample by 111%, effectively doubling its potential scientific payoff. The traditional strategy often waited too long, reaching the final observation without collecting a sample in nearly 52% of the simulations.
“We used a really simple approach that requires very little information, and it still performed really well,” Samuels said.
The findings show how orbital observations and mathematical decision-making could help future robotic explorers make better use of limited time and storage. They also highlight the potential for interdisciplinary research to uncover solutions that might be overlooked when fields work independently.
The researchers are developing a more detailed paper based on the conference presentation and plan to account for how neighboring areas of the lunar surface are related. Adding that spatial information could make the model more realistic and further sharpen a rover’s ability to recognize when it has found a sample worth keeping.
ABOUT THE AUTHORS
Rory Jake Samuels, Ph.D., is a clinical assistant professor in the Department of Information Systems and Business Analytics at Hankamer School Business at Baylor University. His research application areas ranged from bio-pharmaceutical statistics to remote sensing data analysis. His current research includes collaborative projects with researchers at NASA’s Jet Propulsion Lab where his work focuses on advancing methodologies in functional data analysis.
David Dicks, Ph.D., is The Frank S. Groner Memorial Chair of Finance and an associate professor of finance at the Hankamer School of Business at Baylor University. Dicks teaches Advanced Corporate Finance, and his research focuses on corporate governance, executive compensation, innovation, systemic risk, insurance and decision making under uncertainty. His research has been published in the Journal of Political Economy, the Review of Financial Studies and the Journal of Risk and Insurance.
Peter B. James, Ph.D., associate professor of planetary geophysics at Baylor University, served on the science team of three NASA missions: the Lunar Reconnaissance Orbiter (LRO), the Gravity Recovery and Interior Laboratory (GRAIL) and the MErcury Surface, Space ENvironment, GEochemistry, and Ranging (MESSENGER) mission. He is the founder of Baylor's Planetary Research Group, which is collaborating with Goddard Space Flight Center on multiple projects to study the Moon and the planet Mercury. James specializes in the use of spacecraft data to study the crusts and mantles of planets and moons in the solar system. He earned his Ph.D. from the Massachusetts Institute of Technology.
ABOUT BAYLOR UNIVERSITY
Baylor University is a private Christian University and a nationally ranked Research 1 institution. The University provides a vibrant campus community for 20,000 students by blending interdisciplinary research with an international reputation for educational excellence and a faculty commitment to teaching and scholarship. Chartered in 1845 by the Republic of Texas through the efforts of Baptist pioneers, Baylor is the oldest continually operating University in Texas. Located in Waco, Baylor welcomes students from all 50 states and more than 100 countries to study a broad range of degrees among its 12 nationally recognized academic divisions. Learn more about Baylor University at www.baylor.edu.
ABOUT THE COLLEGE OF ARTS & SCIENCES AT BAYLOR UNIVERSITY
The College of Arts & Sciences is Baylor University’s largest academic division, consisting of the School of Earth and Environmental Sciences and 24 academic departments in the sciences, humanities, fine arts and social sciences, as well as 12 academic centers and institutes. The more than 5,000 courses taught in the College span topics from art and theatre to religion, philosophy, sociology and the natural sciences. The College’s undergraduate Unified Core Curriculum, which routinely receives top grades in national assessments, emphasizes a liberal education characterized by critical thinking, communication, civic engagement and Christian commitment. Arts & Sciences faculty conduct research around the world, and research on the undergraduate and graduate level is prevalent throughout all disciplines. Visit the College of Arts & Sciences website.
ABOUT THE HANKAMER SCHOOL OF BUSINESS
Baylor University’s Hankamer School of Business strives to further God’s kingdom through the realm of business, using God-given gifts and academic talents to do so. Faculty and students conduct purposeful research and participate in experiential learning opportunities, all while operating in a Christ-centered mission. Undergraduate students can choose from 13 major areas of study. Graduate students can earn their MBA on their terms, either through the full-time, online or a Dallas-based executive program. The Business School also offers three Ph.D. programs in Information Systems, Entrepreneurship or Health Services Research. The School’s top-ranked programs make up approximately 25% of the University’s total enrollment. Visit the Hankamer School of Business website for more information.