FarmBots

Summary

FSE 104 and First Semester with FarmBots

I had the wonderful opportunity to join FSE 104 as a freshmen under Jared Schoepf. Through that class I was able to learn a multitude of things regarding the engineering process and contributing to innovative decisions and designs. This included learning how to build CAD models and integration of software into physical systems. Through the class, I became a part of the FarmBots team, a project dedicated to making farming accessible to third-world countries. our specific target and stakeholders were based in Suriname, where farming and agriculture is regarded extremely important. With my teammates during my first semester, we were able to develop a problem statement along with a physical representation of the robot whilst gaining valuable insights from our stakeholders regarding specifications of the robot or needs of the farmers. 

Leadership Within FarmBots

By the second semester of EPICS, I was able to take the lead on the software aspect of the project. We developed a object detection algorithm using the YOLOv8 model, a classic model for computer vision and object detection to detect weeds which was accompanied with a mechanical arm to extract them developed by our mechanical team. This experience included working in an interdisciplinary team, pitching our ideas and thoughts to leaders and spending the time and effort on our projects to make it a success. 

Reflection:

Working with FarmBots was the most compelling part of my journey as an aspiring research. It was the first project where I conducted actual literature reviews in computer vision research, talked to stakeholders about their concerns and collected data, processed that data into our prototype, trained models with large amounts of data. I didn't know then but this has become an integral part of my research currently. 

What made this project meaningful was ot just the technical depth, but that research is inherently human-centered. Speaking with stakeholders forced me to think beyond model accuracy and parameters and into questions of usability, reliability and real-world impact. Technology is not built in isolation but for people with specific needs as well as constraints. Conducting literature reviews taught me to situate my work within a broader academic conversation rather than reinventing solutions. Processing real-world data made me analytical and training large models taught me power of compute.

Overall, I believe this project to be the foundation of my current research direction. It was the moment I transitioned from "learning ML" to actually thinking and breathing like a researcher. 

Joy of Living Aspect:

Through this experience, I was able to hone my skills while moving forward in the direction of helping people. This closely relates to what I envision my theme to be; Joy of Living. I was able to learn and adapt our model and robot to the needs of people miles away from me, that taught me the importance of software development process. This project also introduced me to the field of AI and Computer Vision, a career goal of mine and what I am working towards as well.