Category: MeeDm’s Status Reports

I’ve since finished analyzing the data to check if the values are dangerous or not. I’ve also used Tarush’s code and edited it so that if it is dangerous, we send an email to the user. I did a statistical analysis to determine if the eating event was out of the norm or not and then sent emails accordingly (out of 2 standard deviations from the mean).

I’ve also finished up the code for determining the food consumed. I had to work with numpy arrays, etc. to subtract consecutive values and deal with edge cases.

I’ve additionally finished up writing the refresh_botton function that Tarush will call when the user wants. I’m now figuring out the kinks of python goodness with importing and using multiple functions from a file.

I’m also meeting later today with Lucas to figure out the remaining kinks for reading data from tags.

I’m on schedule still, and want to finish up the refresh_button function and reading from the tags.

Tarush have gotten our pipeline to work. We’ve figured out how to pipe data in between the web app and the software portion, and are now working on edge cases.

Lucas and I also managed to make a (crude) load cell design that does what we want it to do (MVP). It hooks up to a custom breakout board, the arduino, and then wired to the pi. We are then able to read in the data and print it out on the pi.

I’m now working on analyzing the data to find the right “danger level” tailored to each cat. Additionally I’m finding the consumed amount by the cat not only the raw weights.

Lucas has since figured out how to put up readable data to the bluetooth, and since I have connected to the tags before to read other data, we are meeting up in lab today to read in the correct data and finish up that end of the pipeline.

I’m well on my way to finishing up the last few things. That is, analyzing the data and then connected to the tags and reading in values. I’m on the right timeline, and that’s mostly all I have left other than edge cases for the project itself.

We demo-ed for this week. Tarush and I have finished figuring out how to communicate between us. I will be looping and checking the tags, and tarush will simply be looking at the files. To make sure this runs smoothly we’ve also implemented mutexes, so that the shared resource (files) don’t have a race condition. We’ve also been working on the characteristics with Lucas, and we finally have a couple done with this.

I still have to finalize the cad design for the load cell since talking to Lucas, but it is ready to print. I didn’t get to a lot of stuff (again), since I also got the covid booster and flu shot all in one go this week and have been sick up until Wednesday. I’ve been pushing myself to work on this, but I will be taking some more of that slack week to get myself at a point to start working again.

I’ve been feeling better, and Lucas and I are meeting today to finalize communications between our parts. By this week the goal is to finish up that portion, to finish up mutexes, and print the load cell.

I finished up tweaking the code for Tarush to then call from the web app to call the tags. While before I had it so that Tarush called my function using a microchip number, I made it so that if he gives the user, I can then update all of the cats for the username. I did this by organizing all the cats informations, and then calling the function to write to the json file. Also before I was making it write to a txt file and I edited it to write to a json file. I’m also working on the cad design for the enclosure. Because the previous enclosure had some design flaws, I’m making it again in blender.

My progress is very on schedule. I’ve been working pretty hard and I just have to wait for the tags to be done. Until that is done I will cad design both the enclosure and the load cell.

I hope by next week the cad designs will be completely done and hopefully when Lucas is done with the PCBs I can then connect to them and figure out 100% the connection to them.

This week I made lots of progress on the actual coding. Tarush and I have already completely decided on how to pass files between each other. I’ve finished writing code that reads in the correct values from a json file (information about the cat and owner) then uses those values to calculate other values such as volume of food/water, which user folder to put the information into, and which cat’s file to put it into as well. I’ve then written all of those values to a text file (probably change later to json file). We’ve also finally figured out how to connect via bluetooth to the demo tag. I’ve been able to take a look at all the characteristics of the tag and can read the values. We have to finalize which characteristics to look at and how to write to it, but that part is not as bad to figure out. I would say I’m on track.

For next week, my goal is to CAD design the load cell and rfid reader, and finish up figuring out how to read and write from the tag and finish up the whole pipeline.

I worked on pretty much half if not more of the design report. Specifically, I wrote up the introduction, design requirements, the main components of the system architecture section, the Tags section of the design trade studies, the system description section, budget and ethical issues. I also formatted the document for the team. I easily spent at least 7-8 hours on this review report.

During class times, I’ve been using bluepy in more detail. Since we switched our raspberry pi, I’ve had to set things up again. I’m figuring out how to figure out how to make the peripheral’s that are a little different discoverable, but I’ve had to put that on the back burner for the design review report.

By the end of the weekend I will have the code ready to connect to the demo tags and send a simple json file to the web app. Honestly I’m still a bit behind on the actual things for the project, and that’s because I had to take a hit with the design review since I didn’t expect to have to do so much work for that.

This week we have been prepping for our design presentation. I went through the slides and outlined all of them in the beginning and then also finalized slides for communication between the hub and the devices as well as the hub and the web app. Additionally I did the diagrams for the web app.

In terms of work on the project, I did research, set up the raspberry pi 3, ordered things that we need, and started writing my code to connect to peripheral devices and connect to them. In terms of research I found some links where others have done similar things using a packet called bluepy, which is a python library that allows you find and connect to devices, then read in packets. It uses bluez, a bluetooth stack that raspberry pi 3’s use. After setting up the raspberry pi, I’ve started a python script to start scanning and connecting to devices. Additionally I’ve ordered some cat bowls, microhdmi to hdmi cables, and a usbc cable to start hooking up our raspberry pi 4.

My goal for next week is to start the multithreading portion of the python script and present for the team.

This Sunday, we did some prep work for the presentation that was on Monday. We finished up deciding on all of our requirements, by thinking about being a cat owner and what they would need. We also polished up the slides, which Lucas presented on Monday. I helped out a with designing the slides as well as input as a cat owner and re-doing the hand-drawn diagrams and was responsible for testing & verification slides, gantt chart, and the division of tasks/labor slides.

We also met up after class to start setting up the raspberry pi and also Lucas and  I started taking a look at the robocat one more time. Because the pieces were messed up, we decided to reprint those pieces.

Lastly, I’ve been doing lots of research this week on how to use the bluetooth on the raspberry pi, and looked at ble5, and how to setup the rasbpi the way I want to. Tarush and I will be finishing up setting up the pi in the lab tomorrow and I will be initializing setup for the bluetooth module as well. I’ve also thought through some hardware diagrams in order to receive interrupts or poll for inputs from the ble device (tags) and make threads to run each of these at once.