UAVOS and Mira Aerospace have completed a flight test of NAVAI vision-based navigation aboard the ApusNeo18 high-altitude pseudo-satellite. During the test, the system maintained navigation under GNSS-denied conditions while operating at an altitude of 16,000 meters, approximately 52,500 feet.
NAVAI is designed to provide uninterrupted positioning for unmanned aircraft systems by using onboard cameras, aircraft telemetry and terrain data as an alternative when satellite-based navigation is unavailable or unreliable.
NAVAI Vision-Based Navigation Tested at 52,500 Feet
The NAVAI system was integrated into the ApusNeo18 HAPS for the flight evaluation.
According to UAVOS and Mira Aerospace, the system maintained precise navigation under GNSS-denied conditions at 16,000 meters.
Traditional unmanned aircraft systems depend heavily on GNSS for positioning and navigation. The companies said GNSS performance can be degraded by:
- Signal jamming
- Spoofing
- Natural interference
NAVAI is designed to provide an alternative by analyzing video from onboard cameras together with aircraft telemetry.
System Combines Visual Odometry and Terrain Navigation
While inertial navigation systems can provide backup navigation, UAVOS said accumulated navigational drift can cause positioning errors to increase over time.
NAVAI addresses this by combining visual odometry with terrain-referenced navigation.
The system compares real-time camera imagery with pre-loaded reference imagery to generate a bounded, non-drifting absolute position estimate when suitable terrain is visible.
According to the company, NAVAI can obtain a position fix from a single image frame at high altitudes where visible terrain features may be sparse.
AI Processes Visual Navigation Data Onboard
NAVAI vision-based navigation uses neural networks and onboard edge computing to process visual information in real time.
Key capabilities include:
- Neural network feature matching: NAVAI identifies distinctive ground features and compares them with an onboard terrain map database to determine aircraft position.
- AI-based noise filtering: Algorithms compensate for cloud cover, haze, low-light conditions and seasonal terrain changes, including snow and vegetation growth.
- Ground Control Station integration: A dedicated interface allows operators to view live overlays of visual matches directly on the mission map.
These capabilities are designed to allow the aircraft to determine its position using visual information when satellite navigation cannot be relied upon.
NAVAI Targets GNSS-Denied UAS Operations
“Testing NAVAI on the ApusNeo platform reflects our commitment to making UAS operations truly autonomous and resilient,” said Aliaksei Stratsilatau, Founder and CEO of UAVOS. “By giving the aircraft the ability to ‘see’ and understand its surroundings, we are eliminating a critical dependency on satellite navigation, which has historically limited drone operations in contested and GNSS-denied environments.”
The ApusNeo18 flight test demonstrated NAVAI vision-based navigation at high altitude under GNSS-denied conditions, using onboard visual processing and terrain-referenced navigation to maintain aircraft positioning.
Source: Press release from UAVOS and Mira Aerospace
No Replies to "UAVOS and Mira Aerospace Test NAVAI Autonomous Navigation on ApusNeo HAPS"