More than 580,000 instances of GPS signal loss occurred between August 2021 and June 2024, according to the International Air Transport Association (IATA), revealing a growing threat to systems that rely on the technology. The Center for Strategic and International Studies (CSIS) further identified communications as a key vulnerability for unmanned systems, raising security concerns as autonomous platforms become more prevalent.
SandboxAQ’s research highlights that many of these systems remain dependent on both GPS and communications, and Ken Devine says, “it is increasingly clear that autonomy depends on software capable of navigating, planning and executing at the edge when communications and centralized command and control are unavailable.”
GPS Signal Loss Impacts Autonomous System Reliability
Autonomous systems require reliable positioning even when global positioning system (GPS) signals are unavailable, a need emphasised by over 480 hours of flight testing conducted by SandboxAQ over 3.5 years. These tests, encompassing tens of thousands of miles and nine aircraft types from drones to military transports, demonstrate the practical challenges of maintaining autonomous operation in signal-denied environments. SandboxAQ’s AQNav software addresses this vulnerability by integrating magnetic navigation with existing systems, offering a means of independent positional understanding when GPS fails.
Unlike systems requiring extensive hardware redesigns, AQNav delivers positioning information designed to complement GPS, inertial and visual modalities. This approach uses quantum magnetometers and artificial intelligence to sense Earth’s crustal magnetic field, creating a navigational signal resistant to spoofing. The company’s strategy focuses on embedding resilient navigation within the existing software architecture of autonomy, rather than demanding a complete platform overhaul.
The potential impact extends to unmanned aerial systems, where minimizing weight and computational load is paramount. AQNav is designed to provide magnetic positioning information without significantly increasing these constraints, a critical factor for platforms already burdened by guidance systems, optics and other essential functions.
According to SandboxAQ, “Building Navigation into the Autonomous Stack” is a key design principle, as they aim to provide a solution that enhances, rather than replaces, current navigational infrastructure. This integration is particularly important as autonomous systems increasingly operate in contested areas or locations lacking robust infrastructure, where reliance on both GPS and communications creates a significant operational risk.
AQNav Software Integrates MagNav for Independent Positioning
SandboxAQ’s AQNav software uses Earth’s magnetic field to provide positioning data, functioning as an independent signal source when GPS is unavailable or unreliable. The system uses quantum magnetometers and artificial intelligence to interpret crustal magnetic anomalies, delivering positional clarity without requiring additional hardware modifications to existing autonomous platforms. Testing demonstrated the value of magnetic navigation as part of a broader architecture, reinforcing the importance of software orchestration alongside sensor types.
Autonomous systems are expected to continuously assess the availability and trustworthiness of various navigation sources, factoring in how planned trajectories affect positioning uncertainty. The integration of independent physical sensing, edge software, and intelligent planning will be fundamental to developing truly sovereign autonomous systems, requiring improved sensors, more detailed maps and sophisticated integrations.
SandboxAQ anticipates that these advancements will allow autonomous systems to maintain operation even when individual dependencies, like GPS or communications, are disrupted, creating a more robust and reliable operational capability. The company states, “Over time, I expect evolving autonomous platforms will continuously evaluate which navigation sources are available, how trustworthy each source is, and how their planned trajectory will affect future positioning uncertainty.”
Waymo AV Testing Validates Multimodal Navigation Approach
Waymo’s experience with over 200 million miles of fully autonomous operation confirms the necessity of multimodal sensing for reliable navigation, mirroring approaches being explored for aerial systems. The company’s vehicles integrate cameras, lidar and radar, each offering unique data and compensating for the limitations of the others, a strategy analogous to the layered redundancy found in aviation, SandboxAQ says. This integration isn’t simply about adding more sensors, but about building a system where the strengths of one modality offset the weaknesses of another, ensuring continued operation even with partial sensor failure.
The parallel with high-definition mapping employed by Waymo is also notable, as it allows onboard computing to prioritize responding to dynamic environmental changes rather than constant position recalculation. By interfacing with existing inertial, visual and other navigational systems, AQNav software can help an autonomous system maintain an independent understanding of its location when GPS becomes unreliable.
This highlights a shift in emphasis from simply collecting data to intelligently fusing it, and emphasises the importance of artificial intelligence in assessing navigation availability and accuracy during mission planning. The aerospace industry’s experience demonstrates that a system-of-systems approach, prioritizing redundancy and resilience, is important for achieving truly autonomous operation in challenging environments.
Source: https://www.sandboxaq.com/post/breaking-the-tether-the-strategic-vulnerability-of-tethered-autonomy
See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.
