AI Takes the Wheel as Perseverance Rover Navigates Mars on Its Own

The Perseverance rover, which landed in Jezero Crater on Mars in 2021 to investigate evidence of ancient life and sample the rocks and another milestone has been achieved by NASA. In December 2025, the six-wheeled rover made its maiden drives on Mars entirely using artificial intelligence laid-out routes. A vision-capable AI examined high-resolution images and terrain data in order to map safe waypoints, enabling Perseverance to travel across Martian terrain on its own.

AI-Powered Navigation

According to reports, during an experiment by the Jet Propulsion Laboratory, engineers had to plan the Dec. 8 and Dec. 10 drives of Perseverance using generative AI. The vision-language model performed the analysis of orbital images and terrain data to detect hazards (rocks, sand ripples, steep slopes) and generate a path with safe station points to the rover. Perseverance then ran the two AI-planned routes, covering 210 meters (246) each. Engineers tested the AI-generated commands in a simulated digital twin of the rover to confirm more than 500,000 variables were correct, and that is why it was safe to deploy commands to Mars.

Implications and Future Exploration

According to NASA, Mars is 225 million kilometers away, which means that real-time joystick control is not feasible; in practice, the planning of rover missions is done by hand using a sequence of waypoints. Administrator Jared Isaacman stated that AI-based planning could help make missions more efficient and increase the returns of science over such distances. JPL engineers say that generative AI can reduce the planning process and ultimately allow rovers to drive at kilometer scales with minimal operator intervention. The development opens the door to smarter robotic systems on the Moon and Mars that will help in future exploration and human missions.