Moscow Aviation Institute (MAI) is developing a unique autonomous navigation system for drones that allows them to operate in the event of a complete loss of GPS signal. This was reported in the press service of the university.
The development will allow the drone to "understand" where it is by analyzing its own "sensations" using an inertial system and computer vision.
A special camera performs the function of vision, and inertial sensors are responsible for the perception of movements. Our algorithm analyzes the incoming information in real time, assessing the position, speed and orientation of the aircraft.
Visual-inertial navigation using the SLAM (Simultaneous Localization and Mapping) method allows the drone to build its own maps of the area and track its position even in difficult maneuvers. The integration of machine learning algorithms allows the drone not only to navigate in space, but also to detect people in real time.
MAI navigation is fully integrated into a self-made flight controller.
This will allow us to adapt the technology for various types of unmanned aerial systems and tasks in the future, such as monitoring remote areas, surveying buildings, industrial facilities and emergency zones.
The system is currently undergoing field tests, and specialists are working to expand its functionality. The project is scheduled to be fully implemented in the second quarter of 2026.
Earlier www1.ru reported that a new technology to reduce noise in the cabins of civil aircraft is being tested by MAI engineers.
