Without a GPS network on Mars, Perseverance has traditionally relied on engineers back on Earth to confirm where it is on the Martian surface. NASA’s newly implemented Mars Global Localization system changes that process. By comparing panoramic images captured by the rover with orbital terrain maps, the technology enables Perseverance to determine its own position in just a few minutes, making longer autonomous drives across Mars possible.
Since landing on Mars, Perseverance has had to navigate without the benefit of a global positioning system. Like previous Mars rovers, it has relied primarily on visual odometry to estimate its position. This technique tracks distinctive geological features in images taken during a drive and measures how those landmarks shift from frame to frame, while also accounting for wheel slippage.
Although this approach works well for short distances, small inaccuracies accumulate over time. With each drive, these tiny errors compound, gradually increasing uncertainty about the rover’s exact position. After long traverses, Perseverance’s estimated location could be off by more than 100 feet.
When that uncertainty grew too large, especially near potentially hazardous terrain, the rover would stop early as a precaution and wait for guidance from Earth.
To correct the rover’s position, Perseverance would transmit a 360-degree panorama to mission operators. Engineers would then manually compare the images with orbital photographs captured by NASA’s Mars Reconnaissance Orbiter. Once they confirmed the rover’s location, they could send updated driving instructions.
Because of the time required for communication between Earth and Mars, this process often took a full day or longer, limiting how far Perseverance could travel before needing updated position data.
New Autonomous Localization System
NASA’s Mars Global Localization system, developed at the Jet Propulsion Laboratory (JPL), significantly improves how Perseverance determines its position on the Martian surface.
Instead of waiting for engineers on Earth to match images manually, the rover can now perform the comparison onboard. The system uses an algorithm that analyzes panoramic images from Perseverance’s navigation cameras and compares them with orbital terrain maps derived from the Mars Reconnaissance Orbiter.
By identifying matching terrain features between the two datasets, the algorithm can determine the rover’s location with an accuracy of about 10 inches (25 centimeters). The entire calculation takes roughly two minutes.
According to JPL robotics operations chief engineer Vandi Verma, the capability effectively provides Perseverance with GPS-like positioning on Mars.
This improvement is particularly important because the rover’s AutoNav self-driving system has already demonstrated strong navigation capabilities. Previously, however, the rover’s driving distance between human updates was limited primarily by location uncertainty.
With Mars Global Localization, Perseverance can pause, determine its exact position, and then continue along its planned route without waiting for instructions from Earth. The system was first used during regular mission operations on February 2nd and again on February 16th, marking its transition from testing into active use.
The development effort began in 2023, led by robotics engineer Jeremy Nash and a team at JPL. During testing, the team evaluated the algorithm using data from 264 previous rover stops, successfully identifying the rover’s location in every case.
How Perseverance Learned to ‘Self-Locate’ on Mars
Ingenuity's Legacy and Technical Implementation
The computing power that enables this capability comes from hardware originally designed for another mission objective.
The Mars Global Localization algorithm runs on Perseverance’s Helicopter Base Station, which previously served as the communication hub for the Ingenuity Mars Helicopter. This base station contains a commercial processor similar to those found in mid-2010s smartphones, making it more than 100 times faster than the rover’s primary onboard computers.
Those main computers were built to survive Mars’s intense radiation environment and rely on hardware technology first developed in 1997.
Ingenuity itself demonstrated the potential of using more capable commercial processors in space. The helicopter was originally expected to complete five flights, but it ultimately performed 72 flights during its mission.
Encouraged by that success, Verma and her colleagues began exploring how Perseverance could take advantage of the base station’s additional processing power.
Implementing the system required careful safeguards. Engineers developed a “sanity check” protocol, where the localization algorithm runs multiple times on the base station. One of the rover’s main computers then verifies that the results are consistent before accepting the final position estimate.
During testing, the team noticed a discrepancy of about one millimeter in the rover’s calculated position. Further investigation revealed damage affecting roughly 25 bits of the processor’s 1-gigabyte memory.
To address this, engineers developed a method to isolate the damaged memory bits during algorithm execution, allowing the processor to operate reliably despite the defect. These technical solutions could also benefit future missions that plan to incorporate commercial processors in space systems.
Conclusion
The implementation of Mars Global Localization marks an important improvement in autonomous navigation for planetary exploration. By allowing Perseverance to determine its own location in minutes, NASA has removed one of the key operational constraints that previously limited the rover’s driving range.
With faster and more accurate positioning, Perseverance can now travel farther between Earth-based updates while maintaining safe navigation across complex Martian terrain. The system also reduces the workload for mission operators who previously had to manually verify the rover’s position.
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.