A collection of my work in robotics, control systems, simulation, and hardware design. Click on any card to view the source code.
Rapid Control Prototyping (RCP) control and system identification, featuring LQR, Sliding Mode Control (SMC) as well as Kalman Filter implementations.
Implemented a Go1 quadruped simulation in MuJoCo by integrating a QP-based convex MPC with Raibert touchdown heuristics for interactive 3D velocity tracking.
Integration of Large Language Models and Computer Vision to automate a robotic arm for playing chess autonomously.
Complete mechanical and CAD design of a pneumatically actuated soft robotic end-effector.
Physical hardware demonstration and testing of the fabricated soft robotic gripper adapting to various object shapes.
Reinforcement Learning applied to Automated Manufacturing Systems for optimized and efficient sorting processes.
Custom differential drive rover integrated with LiDAR sensors, simulated in NVIDIA Isaac Sim and visualized in ROS2 RViz.
GPU-parallelized MuJoCo environment training a Unitree Go1 quadruped via PPO. Leverages NVIDIA Warp and CUDA Graphs for zero-copy state estimation and low-latency physics stepping across 2048 parallel agents.
Model Order Reduction (MOR) techniques applied to simulate and analyze structural damage propagation in U-Bolts.
Complete CAD modeling and mechanical assembly mapping for a custom 4-Degrees-of-Freedom SCARA robotic arm.
Kinematic and dynamic modeling of the custom 4DOF SCARA robot tested within a MATLAB/Simulink environment.
Integration of NVIDIA Isaac Sim and ROS 2 Jazzy. Configured TF trees and Action Graphs to perform real-time SLAM with an AgileX Limo and Lidar sensor.
End-to-End Reinforcment Learning to perform Inverse Kinematic without Hardcoded script.