Skyline Nav AI
Skyline Nav AI, of Cambridge, Massachusetts, builds Pathfinder, software that navigates without GPS by matching camera images to satellite data.
Employee-owned with no venture capital, it was spun out of US Army Research Laboratory technology and has fewer than ten staff. Pathfinder Air serves drones and aircraft, and Pathfinder Edge hardware launched in 2026. Its work is defence-led: a 1.25 million US dollar US Air Force research award and integrations with Kearfott and BAE Systems.
No verified deployments yet
Skyline Nav AI is profiled in the companies register, but no deployment has been verified against published evidence. Records are added as evidence is reviewed and graded.
Products & Solutions
2 productsPathfinder Air
Pathfinder Air is a GPS-independent navigation software solution for drones and manned aircraft.
Aircraft and drones operating in GPS-jammed, GPS-spoofed, or GPS-denied environments lose reliable positioning data, creating mission failure risk. Pathfinder Air provides a resilient navigation alternative that cannot be jammed or spoofed, enabling continued safe operation of drones and aircraft without reliance on satellite signals.
- GPS-independent absolute positioning using computer vision and precomputed satellite imagery
- Seamless multi-modal switching between camera-based absolute positioning and IMU-based relative positioning
- All-weather and all-lighting operation including night vision camera support
- Edge computing deployment - runs on low-power devices including Raspberry Pi 5 and NVIDIA Jetson Nano
5 more features
- No cellular or Wi-Fi connectivity required during flight
- Compatible with multiple camera types including visible light, EO-IR, thermal, and grayscale
- SDK and API available for integration into existing drone and aircraft platforms
- Accuracy within 100 cm, 95% of the time, per LinkedIn company profile and press releases
- Open architecture compatible with ARM, x86, Apple M-Series CPUs and NVIDIA CUDA, AMD ROCm, Intel GPUs
Technical specifications (5)
- Localization accuracy: within 10 to 100 cm (per LinkedIn profile); within 5 m in urban areas, 95% of the time (per contracts page)
- Demonstrated accuracy: 99.9% after 6.2 miles of flight (per T1 news page, March 2025)
- Frame coverage: 100% frame-by-frame real-time absolute or relative positioning
- Edge deployable: runs on Raspberry Pi 5 or NVIDIA Jetson Nano
- Technology Readiness Level: TRL-8 (per LinkedIn company post)
Pathfinder Land
Pathfinder Land is a GPS-independent navigation software solution for ground vehicles. It applies the same computer vision and satellite imagery matching methodology as Pathfinder Air, adapted for land-based platforms.
Ground vehicles in urban canyons, tunnels, underground areas, or GPS-jammed environments lose reliable GPS signal, creating navigation failures for autonomous vehicles, military ground vehicles, logistics fleets, and commercial vehicles. Pathfinder Land provides a resilient fallback navigation layer.
- GPS-independent absolute positioning for land vehicles
- Intelligent switching between camera-based and IMU-based navigation
- Works in tunnels, urban canyons, and areas with GPS signal blockage
- Operates without cellular or Wi-Fi connectivity
2 more features
- Edge deployable on low-power hardware
- Compatible with dash cam and retrofittable into existing vehicle fleets
Technical specifications (2)
- Demonstrated accuracy: 99.98% over 17-mile drive through Greater Boston (per T1 news page, March 2025)
- Terminal error: 6 meters after 17-mile GPS-free drive
Register notes
Notes held in the Aerostack register. Aerostack has not independently verified them.
- GPS jamming and spoofing incidents have surged globally, over 150 aircraft spoofed over Persian Gulf in 24 hours (Reuters)
- GPS signal blockage in urban canyons, tunnels, and contested environments is a critical safety and security risk
- US DoD and allied militaries are actively seeking GPS-denied navigation solutions
- Honeywell, laser ring gyro INS systems (high cost, hardware-dependent)
- Northrop Grumman, precision inertial navigation (defence grade, expensive)
- Emerging computer vision navigation startups
- Software-only approach, runs on commodity hardware, no proprietary sensors required
- Delivers absolute position (competitors often deliver relative only)
- Profitable since 2020 with no VC funding, rare validation of product-market fit
- US military contract vehicles (DARPA ERIS, SBIR) de-risk procurement
Earlier listing summary (3 groups)
Summary text from the earlier Aerostack listing. Figures in it are not independently verified.
- GPS-independent positioning using computer vision and IMU sensor fusion
- Matches live camera imagery against pre-downloaded global satellite datasets
- Delivers real-time absolute coordinates (latitude/longitude): not relative odometry
- Intelligent switching between GPS, camera, and IMU based on confidence metrics
- Runs on-device: ARM, x86, NVIDIA CUDA, AMD ROCm, Apple M-Series: no cloud required
- 10–100cm accuracy across all terrain types and weather conditions
- US Air Force: Pathfinder integrated on next-generation aircraft and drones (SBIR Phase II)
- Kearfott: inertial sensors use Pathfinder to correct IMU drift
- BAE Systems: 360 MVP Sensor navigates GPS-denied areas with Skyline Nav AI
- Winner: 2023 NASA Space Challenge
- Winner: 2025 Best Startup: Inside Unmanned Systems
- Winner: 2025 Pitch the Press (Commercial UAV Expo)
- Winner: 2024 Move America Pitch Competition
- TechCrunch Disrupt 2025 Battlefield Top-20 finalist
- $1.25M USAF SBIR Phase II contract (May 2025)
- DARPA ERIS Marketplace awardee (2025)
Profile data from the Aerostack companies register; deployment records from the verified register. Evidence grades: T1 first-party confirmation, T2 credible reporting. See Our approach for how records are graded.