Link: https://spectrum.ieee.org/drone-gps-alternatives
A recent article in IEEE Spectrum titled "Neuromorphic Camera Helps Drones Navigate Without GPS High-end positioning tech comes to low-cost UAVs" discusses efforts in using neuromorphic cameras to achieve GPS-free navigation for drones.
Some excerpts:
[GPS] signals are vulnerable to interference from large buildings, dense foliage, or extreme weather and can even be deliberately jammed. [GPS-free navigation systems that rely only on] accelerometers and gyroscopes [suffer from] errors [that] accumulate over time and can ultimately cause a gradual drift. ... Visual navigation systems [consume] considerable computing and data resources.
A pair of navigation technology companies has now teamed up to merge the approaches and get the best of both worlds. NILEQ, a subsidiary of British missile-maker MBDA based in Bristol, UK, makes a low-power visual navigation system that relies on neuromorphic cameras. This will now be integrated with a fiber optic-based INS developed by Advanced Navigation in Sydney, Australia, to create a positioning system that lets low-cost drones navigate reliably without GPS.
[...]
[Their proprietary algorithms] process the camera output in real-time to create a terrain fingerprint for the particular patch of land the vehicle is passing over. This is then compared against a database of terrain fingerprints generated from satellite imagery, which is stored on the vehicle. [...]
The companies are planning to start flight trials of the combined navigation system later this year, adds Shaw, with the goal of getting the product into customers hands by the middle of 2025.
What is the benefit of a neuromorphic camera in this application? Looking at NILEQs pages, the application seems to be military fixed wing "drones". As the drone flies forward at high speed, won't the neuromorphic camera flood the processor with events in almost all pixels, all the time due to the forward motion?
ReplyDeleteMaybe there is some optical compensation involved? Or maybe the thresholds are adjusted to only trigger on larger movements
DeleteGPS is not complexe enough as EVS
ReplyDeleteIt makes sense when the GPS is not available. The low cost inertial device has a lot of drift and this drift could be partially compensated by optical image. But from this perspective, a classic image sensor together with an image processing chip may do better job. PixelArt has recycled their mouse sensors for drone stabilization.
ReplyDelete