My Work
From hardware robots to simulation frameworks to deployed software — a collection of things I've built and broken (and fixed).
End-to-end perception system for a bi-manual mobile nurse assistant — multi-camera calibration, depth fusion, object recognition, and shared autonomy on Jetson Nano.
Marker-less 6-DoF pose estimator using vision-only pipeline and multi-view fusion for bottles, syringes, and trays — no physical markers needed.
Cascaded PD and LQR controllers for a physical quadrotor with polynomial trajectory generation. Achieved sim-to-real transfer with under 7mm flight RMSE.
Inverse kinematics and adaptive gait controllers deployed on Raspberry Pi, improving stability on uneven terrain and dynamic balance under perturbations.
LSPB trajectory planning and PD joint control for a 6-DOF manipulator arm. Torque estimation validated within 5% of measured values using motor current profiles.
Full software pipeline to calibrate 14+ sensors and compute Heat Release Rate in real time using oxygen-consumption calorimetry with live monitoring dashboards.
DDPG and A3C implemented from scratch for pick-and-place manipulation on a simulated UR10. Includes imitation learning pipeline for peg-in-hole tasks.
Action Chunking Transformer with VAE latent encoding trained on expert demonstrations. Outperforms behavior cloning by 30% on constrained peg placement.
10,000+ frame synthetic Vision+IMU dataset created in Blender. MSCKF filter + deep fusion network achieving <5% trajectory error, 20% better than vision-only.
Simulation and control stack enabling non-verbal gesture control of a robot in Gazebo. Customizable gesture mapping with 90%+ detection accuracy.
Classical ANMS + RANSAC pipeline alongside HomographyNet trained on MSCOCO. Unified library for high-quality panorama stitching achieving ~1.5px RMSE.
Two-phase pipeline fusing classical filters (DoG, Gabor, Canny) with CNN backbones. DenseNet achieved 92.5% CIFAR-10 accuracy using 30% fewer parameters.
12 FPS dashboard overlaying lanes, vehicles, pedestrians, and collision highlights on Tesla Model S streams. 3D classification, pose estimation, and motion segmentation.
PINN models for predicting object motion under manipulation — incorporating physical priors and engineered features for push/pluck manipulation scenarios.
This site — a fully custom, deployed robotics portfolio with animated carousels, responsive nav, contact form, and Google Analytics. Built from scratch without frameworks.
Personal finance tool that parses PDF bank statements from Amex, Chase, BofA, and Capital One — auto-categorizes transactions with AI and computes exact cash back, points, or miles earned per purchase.
New software projects are in the works. Check back soon or reach out if you'd like to collaborate.
Get in Touch