Mask Background Mask Background

Meet AirTower — a fully autonomous take-off and precision hovering mode for GNSS-denied environments

Hold your UAV in place like it's bolted to the sky—even in jammed, spoofed, or zero-visibility conditions. Whether you're relaying comms over a combat zone or scanning for life in a collapsed building, AirTower gives you unwavering control over your UAV.

Mask Background
99.98%
positioning accuracy
±1m latitude
single point positioning
Fully NDAA
Complaint

Fly the toughest missions without GNSS

Contact

Radio relaying

Hold altitude to extend communication range in signal-blocked or mountainous terrain.

Base defence

Gain a 360-degree view of your perimeters without relying on ground infrastructure.

Tactical ISR

Maintain a reliable vantage point for uninterrupted surveillance in GPS-contested zones.

Industrial inspections

Assess conditions of large-scale assets faster and without putting personnel at risk.

Infrastructure monitoring

Keep an eye on critical infrastructure in remote areas to detect damage or intrusions.

Search & rescue operations

Deliver a steady aerial view to locate victims and guide responders in chaotic conditions.

Autonomous drone navigation, powered by AI Hybrid INS Navigation Kit

Bavovna’s AI navigation kit combines a low SWAP Hybrid INS device with custom-trained AI algorithms for seamless navigation. Trained on chaotic, real-world flights, our algorithms adapt fast and hold steady, even under EW threats or natural interference. All-in at just 800g/1.7 lbs our AI flight control system is ultra-light, plug-and-fly, and ready to mount on any UV platform.

What’s included

Sturdy, EW-protected hardware

Multilayer carbon shell with PEI-doped structure and under-mount for vibration isolation.

AirTower mode for GNSS-denied flights

Precision hover with <0.5% positioning error and no reliance on satellites or maps.

Autonomous take-off, landing, and return to home

Fully automated mission execution with single point positioning of ±1m latitude.

Advanced mission planner

An AI-powered interface, built on ArduPilot to plan and run remote missions

Seamless flight controls

Synthetic GPS output integrates seamlessly into your autopilot system.

Contact
Mask Background

Autonomous hover mode for any UAV platform

AirTower is hardware-agnostic and can be configured for quadcopters, tethered UAVs, VTOLs, and beyond. Whether you’re flying electric, hybrid, fixed-wing, or rotary—our AI INS kit plugs in seamlessly, no structural overhauls required.

Successful deployment for Aurelia X6 Max Pro-D

The UAV completed 30+ km of high-entropy flight maneuvers, followed by an autonomous RTL, all without GPS or map input. The system maintained full stability and sub-5m landing precision despite freezing temperatures and signal interference.

Highlights:

  • Total Distance: 30.88 km / 19.18 miles 
  • Flight Duration: 57 minutes
  • End-Point Accuracy: 4.2 m /13.77 ft (99.99%)
  • GNSS Used: None

Why AirTower mode stands apart

Tailored to your hardware

Easily integrates with any UAV platform and supports custom sensor arrays and payloads, including radio repeaters, sensors, and SIGINT RF modules.

Low SWAP

Runs on a standard +5V supply with just 50W of power draw. Use it together with additional payloads without battery drain.

Composite EMV shielding

Housed in an EMI-shielded enclosure for stable performance in high-interference and extreme weather conditions.

Ultra-long range and high accuracy

Fly up to 62 km/38 miles with sub-meter precision, even in contested or visually featureless environments.

Designed for signal blackouts.
Built for precision.

Contact

Comparison to other non-GPS Systems

Type
Inertial Navigation Systems (INS)
Radio (eg: VOR, LORAN, TACAN)
Landmark (optical)
Magnetic
Bavovna Al-enhanced Inertial
Range
Short
Short
Short
Short
Long
Accuracy
80%
90%
95%
95%
>98%
Nature of errors
Drift, biases, cumulative error of integration
Signal interference and propagation delays
Environmental conditions and landmark changes
Distortion by ferromagnetic, electrical currents and geological formations
Hybrid systems can be affected by the combined errors of the systems they integrate, especially if one system's errors are not adequately compensated by others. Using continuous ML decreases the error rate.