3 ms·
hey HNers, this is sorta related to an idea I have, using a Pi plus an RF transceiver for reading RFID tags. Commercial ones with Bluetooth seem to run $1,000 o
by AlbertCory 4y ago
hey HNers, this is sorta related to an idea I have, using a Pi plus an RF transceiver for reading RFID tags. Commercial ones with Bluetooth seem to run $1,000 or more.
I'm helping out with a food distribution org (lfrad.org). One of their gleaners, Steve, would like an easy way to estimate the weight of the food they've collected from a supermarket. The idea we like is to attach an RFID tag to each box of food (not each item), and segregate the types of food, e.g. a bread box, a produce box, a milk box. Steve already does the repacking, in boxes in his SUV. He knows the average weight of a box of each type of food.
The tags will just say what kind of food it is. They can be reused once they're counted.
Note that the boxes don't come pre-packed from the store, and he doesn't lay them all out on the ground before putting them in his car.
SO: the goal is, he just points his homebrew RFID reader at the car full of boxes, and it counts the boxes of each type of food.
Has anyone ever done anything like this?
He's a former circuit designer, so hardware hacking is not beyond him. I also have other friends who could help. I think a funky-looking homebrew product is actually preferable to a commercial one, since who's going to steal it?
EDIT: thanks for the answers so far. I think some commercial units do work with passive tags at that distance (say, 5 meters), but I'll have to recheck.
The trunk can be open; that's not a problem.
- qbasic_forever 4y agoNFC and RFID doesn't work at that kind of range, i.e. standing multiple feet away. It definitely won't work through a metal car trunk too.
- isoprophlex 4y agoYour current design employs passive tags. Those rfid tags are lit up by the scanner, and thus must be in very close proximity. There's nothing (that i know of) that will report its data passively over that distance. Alternatives could be Bluetooth beacons or something bespoke, but that will require powering the tags... increasing cost and complexity of operation. Edit: you can use computer vision approaches (like qr codes) if you have line of sight to each box. But that's not guaranteed if I interpret your problem correctly) Edit 2: if you do use RFID but position the reader near the trunk opening and scan each box as it enters the trunk, you can keep a running total of the weight added to the car. That might strike the best cost/effort/result balance given these constraints...
- AlbertCory 4y agoEdit window now closed, so an update: After much research and brainstorming with my friend Jerry, I'm hitting on some tags from Chipolo. They're in the same league as AirTags and Tile, designed for consumers to find their lost items, not for industrial applications. Much more expensive than a passive RFID tag, but the "reader" is just your phone, so that's an enormous win. And he only needs a few tags, which will be used over and over.