Team E7: The Wand Makers

Carnegie Mellon ECE Capstone, Spring 2025 — Sharon Lai, Olina Zhang, Nadia Palar

Weekly Status Reports

Sharon’s Status Report for 4/12/2025
Posted by : hsuanjul Sharon's Status Reports

Accomplishments

  • Assisted Olina on CNN model: tested with various waving approaches of the gestures
  • Successfully flashed firmware to new PCB: clone function is now working. Transmitter is able to store IR signal to the external flash and is able to fetch the signal and transmit it. The signal sent out is robust and can be received by appliances. The transmitting distance is closer than we deduced from the data sheet.

Verification

  • Batterylife: turn on the wand without plugging it in. See if the LED for battery is still on after 3 hours.
  • PCB design: the design has gone through design review within team before manufacture. Functioning well so far
  • IR Transmitting and receiving: Tested IR transmission at different distance and angles to the target. The transmission is robust so far with 100% successful transmission within 5 meters and 45 degrees to the target. Failure on transmission largely increase when transmission happens out of this range. I will perform more quantitative test on the transmission later and summarize a more reliable data to compare with our 80% target.

Schedule Update

I remain on track.

Plans For Next Week

Print new wand case for final review.

Olina’s Status Report for 4/12/2025
Posted by : yunjiaz Olina's Status Reports

This week, I’ve been working on figuring out what might be causing the unexpectedly poor performance of our wand gesture recognition system. One possible issue I identified is related to the timer used during data collection. Previously, each gesture was recorded over an 8-second window since there is issue with our code for the timer set up, which may have introduced a lot of irrelevant or noisy data, especially since most gestures take only a second or two to complete. We recently fixed this by adjusting the timer so that each gesture is now recorded for 1.5 seconds instead. Alongside that, we also made updates to the code to support the new timing setup. Right now, I’m testing to see whether this change leads to improved model performance. I am still stay on track, and would finish by next week.

Nadia’s Status Report for 3/29/2025
Posted by : npalar Nadia's Status Reports

Since we have our interim demo next week, I spent this week primarily preparing firmware for both the integration test with the wand chassis and CNN as well as further testing the IR transmission between our wand and the custom receiver. More specifically, this involved doing isolated tests with just the CNN and just the IR transmission as we tried to integrate it on the V2 of the PCB. Though it functions on boards, there are still a few kinks to work out that will be easier to do remotely now that we have isolated the issues and each have the necessary hardware and system knowledge to debug and fix.

Next steps are to continue making the IR transmission more robust both in consistency and maximum range; ideally, we want to be able to demonstrate it working at the range listed in the transceiver datasheet (up to 10m) which will require more thorough testing. Though we had some small roadbumps as discussed in the team status report, I remain on schedule to have the firmware ready to go for Monday and Wednesday’s interim demo next week.

Team Status Report for 3/29/2025
Posted by : hsuanjul Team Status Reports

Risks and Contingency Plans

  • Firmware Compatibility Issues : The updated firmware is compilable on different machines and we were able to successfully flash the firmware to the new PCB. However, functional issues may still occur after further testings. So far we have been able to fix the bugs, but if the plan doesn’t work, we would move back to the original firmware and build upon that.
  • Product quality : The prototype we currently have may break after a lot of trials. We have been collecting data to train model using our version 1 PCB, and noticeable damage has been caused (one of the buttons is no longer sensitive and the battery port is falling off. We are trying to do most of the testing and trainings on version 1 PCB and development board so that our version 2 PCB remains new.

Design Changes & Impact

There is no design changes at this point. We are working on integrating everything and preparing for the interim design review

Schedule Updates

No major changes were made and the schedule remains as is.

Progress / Photos

Fixed wand button: https://youtu.be/jqVDue59yyQ

New PCB (with close up of through-hole for wire connection to IR LED):

Please see the project GitHub.

Sharon’s Status Report for 3/29/2025
Posted by : hsuanjul Sharon's Status Reports

Accomplishments

  • Assisted Olina on CNN model: tested with various waving approaches of the gestures
  • Successfully flashed firmware to new PCB: resolved version conflicts and weird errors across different laptops.
  • Integrated new model: Confirmed that it is safe to reduce buffer size to 7K (7*1024), given that the model size was reduced by a noticeable amount
  • Adjusted the printed wand model: fixed the button problem of easily being jammed and false long press

Schedule Update

I remain on track. Will be able to showcase at least one type of IR transforming with the wand fully assembled during the interim design review.

Plans For Next Week

Perform minor updates to our design and focus on the interim review. Print new wand case for final review if needed. Update and place the receiver PCB if needed.

Nadia’s Status Report for 3/22/2025
Posted by : npalar Nadia's Status Reports

This week was spent primarily fine-tuning and stress testing the firmware on the arrived boards. Sharon assisted with debugging and ensuring that things were functioning correctly for on the board testing. We also collected some data for testing and validation to be used in the final report and presentation.

Development for the custom receiver is slightly behind schedule. Since I was out of town from Wednesday to Friday, I plan to use this weekend to make up progress to be back on track by next week and the interim demo. Firmware for the transmitter is mostly fine-tuned and ready to be integrated with the model. Our primary goal at this point is to focus on system integration for our upcoming interim demo at the end of the month.

Team Status Report for 3/22/2025
Posted by : yunjiaz Team Status Reports

Risks and Contingency Plans
We completed fine-tuning the firmware this week and began integration testing. So far, the system appears to be functioning as expected. Firmware remains the most critical component, but with the recent fine-tuning, we are more confident in its stability.

Design Changes
There are no changes to the design at this stage.

Schedule
We remain on schedule.

Olina’s Status Report for 3/22/2025
Posted by : yunjiaz Uncategorized

This week, I finalized the fine-tuning of the CNN model. I also finalized the input feature set, based on prior experimentation with different column combinations.

To address the trade-off between model complexity and overfitting, I adjusted the batch size and verified that a slightly larger batch size improved training stability without exceeding available memory limits.

Next week, I will move forward with deployment and integration into the system.