SandboxAQ and Acubed Achieve Progress in Magnetic Navigation

Insider Brief

• SandboxAQ and Acubed have demonstrated that their AI-powered quantum magnetic navigation system, AQNav, can achieve real-time aircraft positioning without GPS, even under conditions of GNSS denial and interference.
• In nationwide tests spanning over 150 flight hours, AQNav exceeded commercial navigation accuracy standards, achieving full compliance with RNP 2 and outperforming inertial navigation systems in GNSS-deprived environments.
• The campaign validated AQNav’s ability to operate across diverse terrains, aircraft platforms, and electromagnetic conditions, marking a major step toward scalable, GPS-independent navigation for aviation.

As instances of GNSS denial, jamming, and spoofing continue to impact the aviation industry globally, recent breakthroughs in quantum magnetic navigation (MagNav) from SandboxAQ and Acubed, Airbus’ Silicon Valley innovation center, are bringing the aviation industry one step closer to solving this global challenge and preparing the world for the future of smart automation.

At the heart of this effort is AQNav, our AI-powered MagNav system based on quantum sensors pioneered by SandboxAQ in 2021. Recognized for its innovation, AQNav was chosen from over 120 entries by ACT-IAC for the 2025 Innovations Champion Award and was named one of TIME’s Best Inventions of 2024. AQNav uses advanced quantum magnetometers to read Earth’s crustal magnetic anomalies like a geophysical fingerprint, then employs Large Quantitative Models (LQMs) to filter out electromagnetic interference and precisely determine an aircraft’s position without relying on satellite signals.

SandboxAQ has been testing and refining AQNav since 2022 with Acubed, the U.S. Air Force (USAF), and others. AQNav demonstrated its real-time navigation capabilities in USAF flight tests last July and was recently accepted into the 2025 NATO DIANA cohort, which will further hone its capabilities.

Today, we’re announcing the results from a nationwide initiative with Acubed’s Flight Lab to test the navigational accuracy of AQNav. Meeting the aviation industry’s Required Navigation Performance (RNP) requirements is necessary for the system to be deployed on military, commercial, and civilian aircraft.

AQNav performance was tested across different operational scenarios and showed unprecedented precision. Our shared objective was to understand whether magnetic anomaly-aided navigation could broadly meet commercial aircraft navigation requirements. AQNav’s capabilities clearly exceeded the accuracy required for en route travel between airports—even on our longest flight.

Accuracy
% of flight time
RNP 0.3 (550 meters): 64%
RNP 1 (1,852 meters): 95%
RNP 2 (3,704 meters): 100%

To demonstrate how the real-time capable system would operate in real-world conditions, flight data was collected, reprocessed, and streamed in real time to produce statistical insights, offering representative capability data for joint team evaluation.

Demonstrating Real-World Vigor

What makes these results more impressive is the operational realism of the tests themselves.

SandboxAQ and Acubed focused on designing tests to mirror authentic, real-world aviation scenarios. For example:

• Standard Aircraft Platform: AQNav was tested using publicly available magnetic maps aboard a standard Beechcraft Baron 58—rather than a compensated geosurvey platform. This aircraft was modified only to accommodate the additional AQNav instrumentation. No extensive electromagnetic shielding or specialized noise isolation was used. All sensors were positioned inside the aircraft, powered by AQNav’s software to deliver a clean magnetic signal.
• Use of a Publicly Available Map: For all flights, AQNav researchers used the publicly available North American Magnetic Anomaly Map (NAMAM), which covers the U.S., Canada, parts of Mexico, and surrounding oceanic regions.
• Unfiltered Flight Paths: Flight operations spanned diverse, operationally relevant routes between 200 airports across the entire continental U.S., without filtering based on magnetic anomaly strength, magnetic map quality, or favorable geomagnetic gradients. More than 150 hours of flight data were collected.
• Diverse Geophysical Environments: Data was collected over a full range of conditions, from magnetically rich mountains to sparsely featured plains, reflecting real-world geographies where aircraft might operate without GNSS.
• True Operational Noise: Onboard, AQNav successfully filtered out real-world interference generated by the aircraft, including electromagnetic, vibrational, and other airframe-induced noise.

Elijha Williams, AQNav’s technical engagement manager, said: “Our campaign was not about demonstrating proof-of-concept performance under ideal conditions—it was about proving AQNav’s viability under the noisy, messy, and unpredictable environments real pilots face every day.”

During test flights exceeding two hours, AQNav outperformed the Inertial Navigation System (INS) without GNSS 100% of the time. During a one-hour flight over the challenging mountainous and forested terrain of California, AQNav achieved its best-observed accuracy of less than 74 meters—roughly two-thirds the length of an American football field.

Precision, Scale and Autonomy for a GNSS-Independent Future

This campaign represents a major leap forward on the road to widespread use of AQNav in aviation. By consistently maintaining accuracy across an uncontrolled, unfiltered national testbed, SandboxAQ demonstrated the viability and true operational robustness of AQNav in real-world scenarios. Andrew Sosa Sosanya, a quantum navigation machine learning engineer at SandboxAQ, commented on the importance of this data: “Thanks to Acubed, the USAF and other partners, we’ve amassed a highly relevant MagNav dataset. This creates a flywheel effect: the more data we collect, the faster we improve model accuracy across diverse mission profiles.”

Beyond the significant performance metrics, the campaign further proved AQNav’s scalability, which is critical for aviation’s future. Demonstrating the system’s success on a single flight is an important proof of concept—but scaling is where real progress happens. Acubed’s Flight Lab enables efficient data collection and analysis at scale, accelerating development and validation. This agile platform delivers significantly more data in realistic operational environments, allowing us to mature technologies faster and gain deeper insights into how they will perform on a global scale.

With every hour and sortie flown, SandboxAQ pushes aviation closer to a world where GNSS is just one of many tools used for accurate navigation—and not a single point of failure that jeopardizes lives and impacts commercial and military aviation operations. Through close collaboration with our commercial and government partners, we’re accelerating the deployment of AI- and quantum-powered MagNav systems.

Join us in advancing the future of resilient navigation. By partnering with us, you’ll help accelerate the adoption of AQNav, empowering pilots and operators to overcome the growing threats of GNSS denial, jamming, and spoofing. If you’re interested in collaborating, please email us at AQnav@sandboxAQ.com.