As part of our embedded system class project at the Ulster University we undertook some market research to see if any new avenue of innovation could be found aside from what is currently available for the smart feeding of pets.
Existing smart feeders are monitored and controlled via a software application on a smartphone with set feeding times and schedules. These feeders are purely 'semi- automatic' and are somewhat reliant on the owner. It was found by research that many dogs especially, are divergent in their eating habits and would need a variety of different schedules dependent on their physical and psychological states focused around activity, weight and general levels of hunger.
For example, dogs that are within a healthy weight bracket and highly active may need fed more than just 2-3 times a day. For dogs that are overweight and less active it would be important to monitor how much they are consuming in relation to how much activity they are undertaking in order to control and positively influence long term health. Further analysis and data and be retrieved to determine the mental health of the pet such as sleep disturbances and dips in activity. By monitoring these elements in this particular case, the life of the dog can be extended, vet bills will be reduced and the general level of happiness within the animal will be at it's best, reflecting a positive, more fulfilled and connected relationship with it's owner.
It is found that in order for our proposed feeding system to be fully automatic, the pet’s activity would have to be monitored and communicated in real time to the feeder and the owner. This would allow for the best patterns and measurements of feeding to be given to the pet.
The wearable’s market for human beings is huge and rapidly expanding with activity and fitness monitoring devices being developed and sold all over the world. For the pet industry this would be a new and fresh approach to monitor your pet as well as maintaining a smart feeding schedule. With few wearable’s available, especially those that don’t communicate with a smart feeder, there is a gap in the market.
A set of sensors could be embedded into the collar of the pet, which would relay the data via Wi-Fi to the feeder. An example of the type of data that would be read from the pet’s activity would be calories burnt, distance walked/ran, temperature, sleeping patterns, favourite walking times, and differences in physical behaviour between different types of dogs in different environments. All this data can be used to create the perfect feeding schedule in order to maintain a good diet and peak physical and mental health for the animal.
In order to achieve this proposal and collect the relative data, Farnell Element 14 are assisting us by providing the necessary electronic components of which the WaRP 7 IOT Development board is a key constituent as shown below.
The WaRP7 is built to help developers overcome issues concerning sensor
compatibility, battery life and restricted licences among other things.
A methodological approach will be employed to set up and gain an understanding around the WaRP 7 board which comprises of the following sensors we will be using :
A cardboard prototype of the feeder will be created to facilitate testing and further development of the system which can then be fine tuned to achieve desired results. The current electronics system we have designed for the feeder is shown below where we have used an Arduino MKR1000 to control the various I/Os and O/Ps.
Feeder circuit design consisting of an Arduino MKR1000 board, LCD screen, 2 LEDs, reed switch,
servo motor, phototransistor, pressure sensor and various resistors.
PetBud as a whole collective system will thus consist of :
- Bundle of sensors on the pet's collar.
- IOT server to collect and analyse data.
- Smartphone app / website.
to which WiFi will encompass and connect all of these elements. A block diagram of the system is attached to this post.
Petbud Phase 1 Block Diagram.pdf 104.1 KB