Introduction: IoT Dog Collar

About: Hi, my name is Josh

The IoT Dog Collar gets man's best friend connected!

- Does Fido bark when you leave? Login and find out.
- How active is he? Track his daily activity level.
- Has he been out in the heat for too long? Get a notification on your phone.
- Is he out at night and need a little light to guide his way? Turn on the LEDs from your phone.
- Did he find a squirrel to chase and now you can't see him? Turn on the alarm and track him down.

This dog collar is the prototype from the Intel IoT Hackathon that seeks to achieve all these things and more. If you can see what your coffeepot is doing then you should be able to know what your dog has been up to as well!

Step 1: Connect the Expansion Board

The Seeed expansion board is designed to fit on the Intel Galileo or Edison (with protoboard). Match up the pins and push it firmly into place.

Step 2: Add the Sensors and Buzzer

The collar consists of 5 sensors and a buzzer these need to be plugged into the Seeed expansion board as follows: light sensor, microphone and temperature sensor plug into any of the analog ports (labeled A0-A3); the relay, buzzer and push button plug into any of the digital ports (labeled D2-D8); and the accelerometer into an I2C port.

Because this is only a rough prototype (ie not usable on a dog) the sensors are not attached to the collar directly but if you feel the need these can be sewn on beneath the LED strip.

Step 3: Attach LEDs

The LED strip used for this collar had 3M adhesive backing and therefore were simply stuck into place. If your LED strip does not have adhesive backing then the strip can be sewn on instead. Once the LEDs are in place on the dog collar they will need to be wired to the relay provided in the Seeed kit (and from there to an external power source, ie a battery) since the Intel boards don't provide enough power for most LED strips. The relay can then be connected to the board in one of the digital ports.

Step 4: Adding Code

The code used to power this collar was very swiftly written (in JavaScript) using the Intel XDK. It is not very clean and, quite frankly, pretty confusing. For the sake of your sanity I have not attached here but information on getting all the sensors to talk can be found here on Seeed's Wiki site.