Application Layer

[mlivai]

• [x] (run once) setup functions
• [x] (run all the time) functions
• [x] (run based on conditions) functions

[mlivai]

run once:

• setupCamera --> return into variable
• setupProximitySensor

run all the time:

• function to constantly measure water bowl (using measureDistance)

run based on conditions:

• add pet function
• Calibration function for facial recognition (using takePicture)
• calibration function for proximity sensor (using measureDistance)
• function that waits for a significant change in water bowl and takes pictures/tracks changes (using takePicture and measureDistance)

[ajb974]

run once:

• zeroScale() save zero weight value
• TestScale() save test weight value

run all the time:

• readScale_grams() wait for change in grams

run based on conditons:

• Commit change in food/water to file

[ajb974]

Add generic pet class w/ name, age, fc model, feeding data

[ajb974]

Store feeding data in file rather than program, unique file for each pet.

[mlivai]

incorporate #28

[mlivai]

Some things to still consider for application layer:

• how do we want to store predictions based on images (currently in a separate csv file)
• do we want to store prediction for each bowl? pet?
• do we want to create a separate data file for each pet (if so, the naming of the csv files needs to be adjustable, or we should put each in a unique folder)
• how many pictures do we want to take and make predictions based on every time the camera is called to take pictures and make a prediction (currently a picture is taken for prediction every time a change is detected)

~10 pictures per prediction --> takes the prediction thats most common for the set of 10 and returns that to be stored in the file with food and water change

display info based on pet, but track info based on bowl

• how do we want to signal to the program when to start tracking versus when to pause tracking for calibration (possible with events/other functions)
• add threads?
• others things: how do we want to transfer pictures taken from pi to computer for people to use to train using teachablemachine, how do we want to transfer those two files needed back)