Stereo Vision Baseball Catcher
A gantry robot uses stereo vision to catch a baseball pitch.
This is a conglomeration of various personal, class, or reaseach projects. Research projects often turn into papers so check out my publications page as well.
A gantry robot uses stereo vision to catch a baseball pitch.
For this project, I trained a CNN based on the UNET architecture on a database of labelled cancerous cell images. The goal was to predict cancer with >85% accuracy.
Trained an RNN on State of the Union Adresses to predict sentences given a few letters.
Modeling, simulation, control, estimation, and path planning for a fixed wing UAV. The final result of the project is a simulator built from scratch with implementations of IMU, pressure, and GPS sensors, Extended Kalman Filter and RRT path planning.
This project uses SIFT feature matching and some computer vision techniques to automatically grade scantrons.
A novel way to simulate soft continuum robots in MuJoCo.
An embedded pressure control system I designed for low-level control of pneumatic robots.
SIFT Feature extraction and detection for real-time image replacement.
Our team used a combination of computer vision, controls, and reinforcement learning to autonomously navigate a mini city. Our car is capable of recognizing and responding to signs, stopping at rail road crossings, and staying on the road.
Various implementations of localization and mapping algorithms.
A sensor I developed to give robots a sense of touch over their entire structure.
Baloo is a prototype robot I designed and built to work on whole-body manipulation research.