My group and I designed a device that can easily connect to any machine in addition to a positional system with simple adapter cables to be used to automate the calibration process to find the steering profiles of these machines. The intention was to provide a simple interface that will walk the operator through the process of conducting tests that will automatically be completed by the device, where it will also collect live information about the state of the machine from the positional system and additional sensors on the device and from the machine. The recorded information from all tests can be used to produce a more accurate steering profile of the machine which can be used to improve steering performance of self-driving tractors.
In this project, I explore the meaning of programming a physical button and what it means to think like a hardware programmer. This project was the perfect primer into striking a difference between software and hardware programming.
In this project, I attempted to write firmware for a digital door lock that can accept a sequence of button presses that dictate the user’s eligibility to enter. I put together a simple state machine that outlined the life cycle of the system. A state machine describes all the possible states different components of the door lock can be in and the ways it can reach these states.
The terminal program is to log the time at which a button is pushed. In other words, the exact moment at which a 0 ➡ 1 logical transition is detected from a pin is logged and displayed on the terminal running on a computer.
This project helped me understand how MCUs can display information on an LCD from data gathered using two temperature sensors, an internal and external.
My role in the production was to provide hands-on support to artists at Marvel studios and Disney validating asset performance within Unreal Engine and give feedback on asset topology.
As part of my role at eFuse, I was responsible for designing the functional behavior of the virtual studio package we built out for collegiate esports broadcasts.
After watching The Batman, one scene that stood apart from all the others was the one in which Batman walked slowly toward Penguin after the intense car chase. I came back home and rewatched the scene and I noticed that most of the elements within it could be recreated using CG. So, picking Unreal Engine as my weapon of choice, I set out to do just that.
I took part in the Formula AI Hackathon which introduced participants to three different domains of challenges in the world of Formula 1. The challenge track that I chose was ‘3D Modeling’ which required us to write a blender add-on that can accept JSON files containing {x, y, z} location data of different parts of the track, before being mapped to a 3D model. Out of 212 applicants, my team came 4th!
I used all the concepts that I’d learned from my linear algebra course and applied them to write my own ray-tracing rendering engine. It’s not optimal, in terms of performance, as my sole intention was to be able to see linear algebra in action, in the real world. And to learn more about MATLAB in the process, as I’d just been introduced to it that semester.
