20 percent of methane emissions come from food waste in landfills. Continuous, quantitative montoring is essential to managing and reducing food waste. We developed a data-driven solution that automatically collects and processes compost weights from post consumer plate waste in campus dining halls. Students scrape their plates into compost bins when they return their dishes; our system collects the ongoing weight of the bin without disrupting existing student and staff workflows. These data are visualized and analyzed with our user-friendly graphic user interface, which allows staff to extract insights to inform sustainability goals and operational decisions.

Current prosthetic replacements for mitral valves require extensive suturing that costs valuable time during open heart surgery, especially in cases of Mitral Annular Calcification (MAC), where the tissue is stiff and difficult to suture. To reduce the risks associated with prolonged operative duration, we aim to design a sutureless mitral valve prosthesis that can be deployed quickly and conforms to the irregular, elliptical profile of a calcified mitral valve.

Our design incorporates the Shape Memory Alloy Nitinol and common biomaterials to create a self-expanding deployment mechanism. Additionally, we have created testing models that mimic the shape and texture of calcified mitral annuli.

This project showcases the functional verification and live demonstration of a custom silicon chip implementing a Recurrent Spiking Neural Network (RSNN)—a bio-inspired architecture designed entirely through natural language prompts to a Large Language Model (LLM), specifically ChatGPT-4.
The RSNN was synthesized from Verilog code generated by ChatGPT, validated on tasks like XOR, IRIS classification, and MNIST, and ultimately fabricated through the open-source TinyTapeout ASIC shuttle using SkyWater’s 130nm process.
To verify the fabricated chip, our team developed a complete hardware-in-the-loop test framework using microcotb, a Python-based cocotb-like environment adapted for embedded systems. We also created custom MicroPython firmware for the Raspberry Pi Pico to enable direct communication with the chip for streaming neuron parameters, injecting spike inputs, load test scripts and reading output spike trains in real time.

In collaboration with the Maryland Zoo, we have developed a tracking device capable of monitoring the amount of animal interaction with enrichment devices. Our Arduino Nano, clock module, and SD Card reader allow for our tracking device to monitor the amount of time animals spend interacting with their enrichment devices per hour. Our TailTagger tracker’s electronics are placed inside a 3D printed encasing and attach to any hanging or ball enrichment device. After our Tailtagger is taken out of the exhibit, the SD Card is removed and its contents are uploaded to our interactive Enrichment Tracker app where zookeepers can visualize the amount of animal interaction per hour for each enrichment device. Our TailTagger tracking device is water resistant, animal-safe, and durable – it offers a solution for zookeepers to accurately detect when an enrichment device is being used, which will allow for an enriched animal quality of life.