Drones will fly without GPS using unique fingerprint-based technology

Two navigation technology companies are collaborating to develop a GPS-free navigation solution for drones.

Advanced Navigation has partnered with global defense company MBDA to jointly develop a fingerprint-based navigation system that integrates NILEQ positioning technology. This writes interestingengineering.com.

The technology uses advanced neuromorphic sensors to create and compare terrain “fingerprints” and also records terrain changes as the drone moves over the ground, comparing the data with an existing database of the Earth's surface.

The visual navigation system improves inertial navigation by providing accurate location updates at regular intervals, allowing for location recalibration.

Conventional UAVs use high-resolution cameras that provide huge amounts of data that must be compared with satellite imagery using computationally intensive methods. This creates difficulties for small vehicles with limited power, such as drones, where it is often not possible to integrate such computing resources. The NILEQ system uses a neuromorphic camera modeled after the human retina to overcome these drawbacks. A neuromorphic camera produces significantly less data and operates at a much higher speed than traditional cameras that take sequential images. Instead, it detects changes in brightness on an individual pixel basis.

200% Deposit Bonus up to €3,000 180% First Deposit Bonus up to $20,000

This efficient data generation enables real-time processing. It uses proprietary algorithms that generate topographic footprints that correspond to the vehicle's position. The terrain images obtained from the camera's raw data are compared to a preloaded footprint database created from satellite imagery. The footprinting process compresses the data, allowing the database to be stored compactly on the vehicle. This design minimizes the computational requirements during real-time search and comparison, making it suitable for drones and other resource-constrained platforms.

Satellite navigation is often used to determine location, but it is not always reliable. To provide a single, reliable location readout, Advanced Navigation combines the output of a neuromorphic camera and an ANN using AI-based sensor fusion software. This method allows the system to operate similarly to a GPS signal, making it easy to use without the need for extensive user training.

By reducing the data load and computational requirements, this system offers a practical and efficient solution for GPS-free navigation in various environments, increasing the capabilities and resilience of drones.

Due to the limitations of operating in the infrared range, neuromorphic cameras can only be used at night. However, infrared-capable versions are being developed and should be available in a few years.

While neuromorphic cameras are more expensive than traditional ones (typically around $1,000), they can be used with less expensive inertial navigation systems (INS) to provide a more cost-effective solution.

The integrated navigation system will undergo flight testing later this year and should be available to consumers by mid-2025.