Portfolio
A curated collection of my research and design
Zero-Shot LLM-Driven Multi-Robot Coordination
A framework enabling users to orchestrate a fleet of sensor-less, differentia...
Residual Reinforcement Learning for VQ-BET Fine-Tuning
Improving grasp accuracy of a pre-trained VQ-BET base policy by layering a re...
PPO Reinforcement Learning for Quadruped Locomotion
Trained a robust locomotion policy for the Unitree Go2 quadruped robot within...
Action Chunking for Imitation Learning
Enhancing VQ-BET performance via action chunking, hyperparameter sweeps, and ...
AI-Enabled Mobile Robotics Platform (VLM Servo)
A mobile manipulator integrated with a multi-layered, AI-driven software stac...
Generalizable Imitation Learning for Domestic Tasks
High-performance computing pipeline using LeRobot to train Action Chunking Tr...
VQ-VAE Hyperparameter Optimization for Bimanual Policies
Troubleshooting and optimizing bimanual robotic policies on the YOR (Cone-E) ...
Low-Cost Software-Stabilized 3D LiDAR
A highly functional, cost-effective, and perturbation-resistant LiDAR system ...
MPC-Based Quadcopter Control using SQP
Designed a Sequential Quadratic Programming (SQP) solver from scratch to driv...
Neural Network Solutions for Inverse Cauchy Problems
A multi-layer ANN designed to solve a 2D elliptic PDE for detecting material ...