NASA completes first AI-planned drive on Mars with Perseverance rover
Updated: Feb 2nd, 2026
 |
Image: NASA |
Instead of human operators planning every move, NASA used a type of AI called a vision-language model. The AI studied images and terrain data from orbit to find hazards like rocks, boulders, and sand ripples, then created safe routes for Perseverance to follow.
“This demonstration shows how far our capabilities have advanced and broadens how we will explore other worlds,” said NASA Administrator Jared Isaacman in the statement.
“Autonomous technologies like this can help missions operate more efficiently, respond to challenging terrain, and increase science return as distance from Earth grows,” he added.
Before sending the commands, engineers tested them on a virtual version of Perseverance, checking hundreds of thousands of data points to make sure the rover would be safe. With AI-generated waypoints, Perseverance drove 689 feet (210 meters) on December 8 and 807 feet (246 meters) on December 10.
“The AI helps the rover see the terrain, figure out where it is, and plan a safe path,” said Vandi Verma, a JPL roboticist. “This will let rovers travel farther while reducing the workload for human operators and even flag interesting rocks or features for science teams.”
NASA hopes this technology will play a big role in future missions to the Moon, Mars, and beyond.
