(no title)
AShyFig | 1 year ago
The drones I fly for scouting are on various premade flight plans. DroneDeploy.com offers a good application for mapping fields from 300' high, and FlyLitchi.com works well for custom paths (flying to one spot in the field, dropping down low to take a high resolution picture of the crop, then zipping back up and repeating several times.) I can't see much benefit to full AI control of the aircraft. The current mode of operation has them on preset "rails" with room for adapting to obstacles using ultrasonic sensors. I'd like some visual adaption capabilities, perhaps something to do with lidar and SLAM (Simultaneous localization and mapping), but my drones don't have these sensors, and the last time I looked into this a DIY solution was out of my wheelhouse.
I've only seen water trials of spray drones, but my understanding is the lower the better, within reason. depending on the vortices the aircraft generates the spray booms need to be about 4-6' off the canopy top, so ~15' total flight height. You can find a little bit of an overview here: https://sprayers101.com/drone-sprayers-are-we-ready/
dogcomplex|1 year ago
Sounds like a lot of legal overhead is mainly what's standing between drone spraying... 30-60 day delay between plan submission and operation, not easy to keep that practical. Though at least yeah - 15' seems like a much more manageable height - hopefully that would at least be underneath any neighboring aircraft
Fair on the LLM - always use traditional algorithms if available and with effort/time for such things. LLMs are just the convenient duct tape / superglue to throw at handling edge cases with some confidence they'll generally use common sense about it. Untrained intern equivalent which can be combined with the algorithms if they're brittle.