Go Down

Topic: Low-Cost, Long-Range (30m+) 3D Direction Finding (Read 347 times) previous topic - next topic

I have a drone, and what I want is to follow a beacon that can be as far as 30+ meters away.  I need a nudge in the right direction -- namely, which tech I should be using.  Thank you for any help!

Some [rough] requirements:
- Keeping cost is essential for me! (looking for ~$100)
- Only direction in 3D is required, not distance
- Doesn't have to be super accurate; within a meter is fine
- Doesn't need to be 100% robust, though it can't be too flaky
- Only need to track one object
- Tracker size would ideally be less than 6" diameter (not including antennaes)
- beacon needs to be small, less than 3" diamater

Some tech I've looked at but having trouble finding the one to dig into:
GPS - seems like it's probably the best, though I worry about accuracy. can the beacon be small enough?
RFID - active RFID seems promising, but the tracker would probably be way too expensive
Bluetooth low energy - seems great, but I worry about the range
RDF - I think the array would be too big? not sure about accuracy either
RSSI - worry about accuracy

thanks again for helping point me in the right direction here with the tech scheme.

PeterH

I don't think it would be feasible to develop a DIY solution with 1m resolution using any of those technologies.

If it's in a controlled environment you might be able to detect a modulated IR beacon using a scanning directional IR receiver. If your 'drone' is an aircraft then I doubt that would be feasible, but if it's a ground-based drone it might be. Otherwise, you could use an IR camera (or an ordinary visible beacon and ordinary camera and use image processing techniques to locate the beacon. The image processing would require a PC or something - a microcontroller would not be a suitable platform. An embedded PC or an RPi or similar would probably be up to the job.

Another possibility is to use an audio beacon and compare the timing from multiple receivers to determine the bearing to the source. How feasible this would be would depend on what environment you're working in - for example, I'd expect an aircraft to be quite noisy and that might make this approach impractical.
I only provide help via the forum - please do not contact me for private consultancy.


Go Up