Carnegie Mellon ECE Capstone, Spring 2024: Max Adams, Derek Kim, Sid Sapra
There are no significant risks, the IR sensors have been our bigger concern and so far they seem to be working well. The supply voltage was a little funky, but adding a voltage regulator should help. We are looking to set up both the people counter and a bench sensor module in the UC gym this week.
No changes have occurred this week, and our schedule remains the same.
This week, I finished the first implementation of the backend API. I also worked on writing code for the compute units on the sensor modules to be able to communicate with the backend. I mostly worked on testing and debugging the backend API by using Postman to send requests. Additionally, I changed the security settings and other parts of the AWS EC2 configuration to better fit our needs. I am on track with my schedule with my tasks. For our team, we plan on finishing up putting the sensor modules together tomorrow. Once the sensor modules are operational, the rest of the week I plan on testing communication between our sensor modules and our backend server.
This week, I initially set up some testing for the IR sensors in both the people counter and the bench sensor modules. The supply voltage was a little unstable, so I’m looking to add the 5V voltage regulator tomorrow to help with that. I also helped Derek with some testing with the EC2 instance as we look to set up everything this week in the gym.
Our progress is on schedule, it is essential for this next week that we have the people counter and the sensor hardware set up in the UC gym so that we can begin collecting data. I do not have any immediate projects this week so I look forward to having plenty of time to work on finalizing everything.
As of now, there are no significant risks in the project. We look to build our people counter and a bench sensor module this and next week, and have everything set up in the gym so that we can begin collecting data for our predictive model.
There were no big changes to the system as a whole. We have decided to focus more on the model using the people counter component along with a bench sensor set up in the gym. We look to set these up this week.
There was a slight change to our schedule with setting up the components and testing certain pieces.
Part A was written by Max Adams
While times have returned to normalcy, our product design applies to the importance of social distancing and crowd control. Our product solution addresses the need for social distancing by providing users with a mobile window and predictions on the crowding of the UC gym. While our product is specific to the UC Gym, it is scalable to other gyms. With real-time data, governing bodies can ensure that gyms are abiding by occupancy limits to ensure public health and safety are protected.
Part B was written by Sid Sapra
Our app offers a comprehensive solution to address the increasing need for efficient gym utilization, aligning with cultural values of time optimization and health consciousness prevalent in many societies today. By providing real-time tracking of gym occupancy and predictive analytics on crowdedness and machine availability, our product empowers users to plan their workouts effectively, optimizing their time spent at the gym. In cultures where time is a precious commodity and adherence to fitness routines is highly valued, our app becomes a valuable tool for individuals striving to maintain their health and fitness goals amidst busy schedules. Additionally, by fostering a sense of community and consideration for others’ workout experiences, our solution promotes a culture of shared responsibility and respect within gym environments, enhancing the overall experience for all users.
Part C was written by Derek Kim
Our system addresses excess carbon emissions by giving gym goers and gym owners a medium of tracking crowding and the real-time occupancy of a gym. At times when there is very low crowding, gym owners can save resources by reducing electricity use to lower their carbon footprint, helping the environment and saving money. Also, gym users who drive only need to travel to the gym at times best for them if there is low crowding. If a gym is very crowded users can opt for other activities or at-home workouts, thus reducing their carbon emissions.
This past week I began testing the active IR sensor by setting up a circuit with a voltmeter and seeing how different distances were affecting the output. I also worked on my team’s design report. We finalized the bulk of our orders and looked to begin testing this upcoming week and building our components.
We are on schedule, but we want to hastily set up the people counter and a sensor module to begin collecting data at the UC gym.
This upcoming week I’d like to have integrated the IR sensor with the NodeMCU for the bench sensor module. I would also like to finish some formal testing with the sensors themselves.
This week I ordered the NodeMCUs and IR Sensors for the Sensor Module. They finished arriving yesterday, and I have begun to set up a testing space using an old 18-100 toolkit (see attached). I am currently having issues connecting to the NodeMCU as my computer is not recognizing it. It is most likely a cable issue. I also worked on our Design Report which is due this upcoming week and on our Design Presentation which Sid presented.
Today and next week, I look to finish setting up the NodeMCU (connect to my computer and flash software onto it). I also look to begin testing the IR sensors and hopefully have them working with the NodeMCU towards the end of the week. I will also contribute to our design report, which is due on Friday 3/1.
I am currently on schedule according to our chart.
Materials:
This week, I was able to finalize an overall list of components for my part of the project. Currently, the sensor module will consist of a battery pack, a NodeMCU, and an active IR sensor with a 100-550cm range (within what we need for our use case). I was debating trading the IR sensor for an ultrasonic sensor in our design for the sensor module because an ultrasonic sensor module was said to be more reliable from my research and is also much cheaper per sensor. Still, the UC gym is very noisy and would very likely interfere with it. I also concluded that it would make me more susceptible to outside activity because of its wide detection field. (I was looking more closely at the HC-SR04 for ultrasonic) . I also brainstormed a potential detection algorithm for the NodeMCU software with the IR sensor. This is something I’d like to test and continue to develop when I receive the sensor so I can abide it by the sensor.
My progress is on schedule according to our current chart, but I would like to order my components this week. My main concern is the IR sensor, and ensuring that it abides by the range of 100cm-550cm. I’d like to also finish our design presentation and will be working on the report this upcoming week.
I worked on the proposal presentation with my teammates and researched possible components for our idea. I am looking at how we could get wifi on a Jetson, and I have also been refining areas where we can set up both our sensors and camera module. I am currently on schedule, and in the next week I’d like to have our design flushed through and some parts ordered. I also look to finish talking to the gym administrators and get a final answer from them.
As of now, our most significant risk is not being able to set up our camera in the UC gym. We plan on meeting with gym administration this upcoming week to settle this. In our meeting, we plan to introduce our project and our intentions with using a camera. In the case that they reject the camera, we have a few alternative solutions for the role that it fills. (Collecting data for predictive models)