Is anyone already working on the PetFinder use case?


#1

Hi all,

I saw as possible use case that TTN could be very well used for tracking lost pets ( like cats and dogs) especially in dense urban areas.
But is anyone already working on these use cases?

It seems a great idea and Im thinking of starting something up, but wouldnt want to create too many duplicate use cases.


Best practices when sending GPS location data
Is anyone trying to make tracking collars for wild animals?
(Brendan O'Gorman) #2

@ArjanS - Consider some of these companies are further ahead in that game, on the same path, however some thoughts;

http://www.wareable.com/internet-of-things/the-best-pet-wearables-trackers-and-gps-for-dogs-cats-and-more

  • Price seems to be an opportunity.

  • Getting heard, with many companies ahead you would want substantial product differentiation and marketing strength (big resources, ability).

  • Half the world wakes up brainwashed every-morning - you could probably launch a kickstarter and many would think that is an excellent new idea.


(Flavio Maeda) #3

@ArjanS Have you seen www.clickey.eu ? They have a LoRa ready tracking device that can be used for pets and we are thinking on using it.


#4

You also have trace.me working on lora tracking. Not a easy task making a small gps device not power hunger. The lora tracking is not implemented in LoRa yet. Havent seen a shop ready product yet, one could find in a petstore.


#5

Thanks to all for the links. Had not yet seen most of them.

@Brendan the product in the article seem focused around GPS systems. High monthly cost, low battery life. Seems like LoRaWAN would be perfect creative destruction in that market. Of course, risk remains if those competitors move into LoRa
@flavio clickey.eu indeed uses LoRa already but they mention wanting to use the KPN network and also aim to ask for monthly fee's. I hate being charged monthly fees and I suspect majority of pet owners will agree. What do you think?
@Fokko trace.me is working on LoRa tracking, but according to their website they seem to be focused on creating the electronics parts for it. Not so much as to developing a B2C application. Perhaps it may be an idea to team up with them, since they are also in NL. Do you know them personally?

I think that creating a low price, free of monthly charge, nicely designed (small and light), accurate positioning tracker for cats/dogs/kids/bikes/whatever which LoRaWAN can offer will completely upset this market. From what I read in user reviews now most current products fall in the category "nice try, but not quite there yet".


#6

No I don't know them personally.

low price, free of monthly charge, nicely designed (small and light), accurate positioning tracker ...."nice try, but not quite there yet"

Don't underestimate the making of such a thing. As it seem the technology seems to be not there yet. All the Lora network inc. KPN, Mobistar and Thethingsnetwork e.g. just rolled out. Hardware to make nodes is coming now. So if you want to make trackers for cats/dogs/kids/bikes/whatever this is the time.


(Flavio Maeda) #7

@ArjanS. I agree with you. This is something that I definitively would like to invest. Pet market in Brazil is huge so let me know if you want to cooperate on this.


(Patrick) #8

MediaTek has been working on a proto for a real-time bike tracker. Warning: not LoRa and requires PubNub and Mapbox accts. Might could be repurposed...
GitHub Repo >


#9

@flavio Cool. I'll keep it in mind. I will try and get some thoughts on paper and get back to you.


#10

I understood that Peerby became one of the founding members of TTN, just because they want to be able to track what is lend. They ony have "verhuishondje" at the moment, but same usecase applies imho :relaxed:


#11

Hi Fred, Im not sure if I agree. Yes, the underlying principles are the same. But creating the application is very different.

For example, fixing a node to a dog will require a collar. Fixing a node to a loaned out item like a chainsaw or verhuishondje will require a completely different way of attaching node to chainsaw. Also the way an acompanying app will be set up requires a different way of thinking.
How accurate must positioning be? Must it work indoors? How large may the total node be? How heavy? GPS on continously? Upon request?

Right now were struggling with such questions...


#12

@ArjanS: thanks for the link to trace.me -- they seem to have some really interesting products in this space! I've contacted them for some quotes and it'll be interesting to see what they say.

All: what is involved in making these devices (e.g. the ones from trace.me) work with TTN? Is it reasonable to assume that the devices would come with the option of specifying a prioritized list of networks for which the device should attempt to connect to? Or would they typically only have a single network? I've seen the notion of a network key being used, but I'm fuzzy on whether this applies here or if it's just for communication between gateways, routers, etc. Also, is it fair to expect there to be some kind of fallback option, i.e. try TTN first, then if that doesn't work try the next on the list, etc? I wouldn't be opposed to relying on a commercial network in the event TTN is unavailable, but of course that would be hard if every device required an activation and maybe infeasible if each activation required a substantial subscription fee.


#13

the "fall back" option you describe is being discussed on a technical level now elsewhere on this forum: https://www.thethingsnetwork.org/forum/t/kpn-komt-achter-ttn-aanhobbelen-samenwerking-mogelijk/472/17


#14

A short update on this use case.

We elaborated some more on the concept and created a website http://trackey.net/.

Currently, we are prototyping with GPS and a teensy based node. Very much awaiting the coming of the TTN gateways since that will make life (and testing) so much easier!

@flavio if you still want to cooperate on getting this petfinder to the market, lets set up a call to discuss!


#15

Have you given any thought to weight of the tracker?

I have been asked to work on a tracker for ...hedgehogs! Weight is an issue - it needs to weigh <10g

I wondered how heavy your tracker is likely to be?


#16

Weight is an issue certainly, but also think about durability, waterproof, etc.

One of our main learnings is that your target animal determines what you can do with it. Our main target now (black labrador that runs away frequently and is impossible to find in the dark 😄) can handle some more weight than for instance smaller dogs or cats.

So were starting out with some more weight and gps, but may shift to smaller units and just triangulation if that works satisfactory.

Cool project btw, on hedgehogs. What is it about?


#17

Here's the link about the hedgehog challenge: http://www.smartearthnetwork.com/pages-104

I sympathise with the escaping labrador issue - we had one when I was a kid that would go on its own adventures. The pet tracker is a great idea - people are fascinated with their pets.


#18

Nice! How are you planning to fit it to the hedgehog, can you use a collar?

One of your problems will be how to conserve battery and still give out a constant tracking of gps signals. Is the GPS a necessity? If the hogs are just in that park, getting enough gateways around it may increase location accuracy sufficiently just using lora triangulation.


(Flavio Maeda) #19

Hi @ArjanS. That sounds very good! I will PM you with my Skype ID.


#20

They get glued onto the spikes! Luckily they aren't too fashion conscious!!

I think you are right about gps - the lightest gps module I've got weighs 9g on its own so I'm hoping that triangulation may be sufficient - if my maths is good enough :wink: