Weekly Status Reports

Nadia’s Status Report for 3/15/2025

The PCB arrived! In light of this, I spent this week making sure that the firmware could flash on to this PCB as well as performing some preliminary hardware tests. I also started laying down the groundwork to add the custom receiver functionality. Since we have multiple copies of the PCB, we are also ordering an additional ST-link debugger so that Sharon and I can both work on verifying the firmware remotely.

I remain on schedule. For next steps, I will continue to test and develop firmware on the board itself to verify it works as expected on the PCB. Additionally, we are looking to order a IR transmitter so we can further develop the custom receiver modules.

Team Status Report for 3/15/2025

Risks and Contingency Plans

Although there haven’t been a lot of major changes to the updated PCB, we might still encounter bugs when integrating it with the new PCB. At this point the most vulnerable part of the project is the firmware, and Nadia and Sharon are devoting most of the time to it. The old version of firmware is rather stable, so in the worst case where we are not able to completely debug the new firmware, we would make minor modifications on the old firmware.

Design Changes

No changes to our design at this point.

Schedule

We remain on schedule.

Sharon’s Status Report for 3/15/2025

Accomplishments

  • Received manufactured PCB: did basic power testings on the board to make sure there is no short circuit on the board and all pins are soldered on the correct pads. Pulled test points to high and verified basic function of the MCU.
  • Started testing the firmware: flashed the updated firmware to the board. Assisted Nadia with debugging the firmware.

Schedule Update

I will spend most of my time assisting Nadia with the firmware for the rest of the semester.

Plans for Next Week

Assist Nadia with firmware testing.

Olina’s Status Report for 3/15/2025

This week, I spent time refining and testing the CNN model. I employed a series of Conv1D layers with LeakyReLU activations and a final Dense layer for classification. I spent time adjusting the hyperparameters and experimented with various learning rates for the Adam optimizer. Through experimentation, the optimal balance between convergence rate and stability was achieved with the learning rate of 0.001. Also, I experimented with different combinations of columns as input features and decide on a better choice for the combination.

There is a trade-off when we need to balance model complexity and overfitting. I found that larger batch sizes have better training stability, but they require more memory in order to cache intermediate results during training.

Next week, I will perform the final fine-tuning of our model.

Sharon’s Status Report for 3/8/2025

Accomplishments

  • Last week was midterm week and everyone on the team were extremely busy. Therefore, we decided to take a break from the project.

Schedule Update

PCB verification part is a bit off track since we are still waiting for it to be manufactured. I will work on verification as soon as the board arrives after spring break.

Plans for Next Week

Assist Nadia with firmware testing. PCB verificationb

Nadia’s Status Report for 3/8/2025

I spent the bulk of the week leading up to spring break working on the design report due on Friday. I was busier than expected during spring break itself, so not much progress happened during that week. I am continuing to verify and test code on the breakout setup to prepare for flashing onto the board. Currently, the biggest concern on my end is making the code as reliable as possible and ensuring that the user control handling is robust.

I remain on schedule. The PCB is probably going to arrive soon, so I will coordinate with Sharon to come up with a test plan since she will also have to verify the PCB initially once it arrives. Until then, I will work more extensively testing on the breakout/breadboard setup to minimize potential issues during PCB testing and system integration.

Team Status Report for 3/8/2025

Risks and Contingency Plans

Our biggest concern at this point is ensuring we can successfully integrate before the interim demo. To mitigate this, we will extensively test our individual parts to minimize the amount of sources of error before attempting to integrate onto our custom hardware in the coming weeks.

Design Changes

No changes to our design at this point.

Schedule

We remain on schedule.

Part A (Olina Zhang)

Our product addresses global needs by providing a simple, accessible solution to interact with IR-controlled devices, offering an affordable and user-friendly option without the need for sophisticated or expensive smart home infrastructure. The processes to clone and transmit signals are simple and easy to learn, and the use of clear gestures makes it ideal for non-technologically savvy users. In this way, our wand can bridge the gap for people who may not be familiar with traditional remote controls or more complex voice-activated systems.

Part B (Nadia Palar)

In a cultural sense, our wand is designed to be easily used since it does not inherently rely on languages or symbols specific to any culture. The gestures themselves do not have any specific intonation or meaning since they were chosen chiefly for their optimality with regards to detection instead of any cultural meaning. Further, the wand’s simplicity in design is also intuitive enough for individuals who may not have prior exposure to high-tech gadgets.

Part C (Sharon Lai)

Our wand offers a more environmentally friendly approach than typical remotes because it uses a rechargeable battery. Typical remotes for IR controlled devices use disposable batteries which in turn constitute electronic waste. In a global context where environmental sustainability is becoming increasingly important, our product solution contributes to minimizing the carbon footprint associated with the overproduction and disposal of electronic products, offering a cleaner alternative to traditional methods of controlling devices.

Olina’s Status Report for 3/7/2025

These 2 weeks, I compared the CNN, RNN, and LSFM models. While all three models performed well in training, CNN demonstrated the best generalization with minimal overfitting, the fastest training time, and high robustness to noise. In addition to the comparison, I worked on fine-tuning the models to improve their performance and reduce overfitting, focusing on optimizing hyperparameters and refining preprocessing techniques.

During fine-tuning, I encountered difficulties in balancing model performance and overfitting. Additionally, optimizing computational efficiency while maintaining accuracy was challenging.

Next week, I will complete the fine-tuning process and finalize model evaluations. Additionally, I will analyze the impact of fine-tuning adjustments and document findings to guide further improvements.

Nadia’s Status Report for 2/22/2025

This week, I got started setting up some physical tests (loopback, basic communication with the peripherals to ensure working) as well as finishing porting over the preliminary code to FreeRTOS. More generally, using an RTOS will make having concurrency between the user-facing tasks (such as status LEDs, mode switching) easier to implement. I also reduced reliance on a debug terminal with serial since the debugger within the Cube IDE provides many graphical interfaces for using gdb. For deliverables, this also involved documenting the overall flow for the design document due on Friday.

I remain on schedule. Next week, we will focus our attention on finishing the design report since it is due this Friday, and I will begin tests to ensure that we retain full functionality when switching to an RTOS based setup. I aim to have this completed by the end of spring break so we can do more integration tests as well as start evaluating the ordered PCB.

Sharon’s Status Report for 2/22/2025

Accomplishments

  • Wrapped up my part for a firmware update and handed it over to Nadia.
  • Made final adjustments on BOM and layout for PCB, and ordered through JLC. It is expected to ship next week. We are looking to receive it by the end of next week or at the start of the week after next.
  • Made adjustments on 3D-Model: put an additional hole for an LED signal on the wand body to align with the updated PCB for better usability.

Please check the project GitHub for updated details.

Schedule Update

PCB verification part is a bit off track since we are still waiting for it to be manufactured. As we wait for the PCB, we will focus more on the firmware side so that we will be able to have a working firmware once we get the customized PCB.

Plans for Next Week

Assist Nadia with firmware testing.