Caterpillar
Robotics Engineer
I develop and deploy production-grade positioning, estimation, and control systems for industrial machines operating in real-world environments. My work focuses on building kinematic propagation and sensor fusion frameworks that combine GNSS and IMU data to achieve millimeter-level end-effector accuracy. I design modular, platform-agnostic autonomy algorithms and validate them in high-fidelity simulation environments before deploying to production hardware.
C++
Python
Kinematics
Control Systems
Simulation
Protobuf
Maryland Department of Transport
Robotics Software Engineer
Developed an end-to-end autonomous inspection system for storm-drain pipelines using aerial and ground robots. Led efforts across localization, perception, and mapping in GPS-denied environments, implementing VIO with EKF-based sensor fusion to achieve reliable navigation. Deployed a real-time defect detection pipeline using YOLOv5 on NVIDIA Jetson Xavier (45 FPS) and built a 3D reconstruction system using Gaussian Splatting for accurate scene modeling in feature-scarce underground conditions.
Python
ROS
Gaussian Splatting
State Estimation
UMD Robotics Research Lab
Robotics Researcher
Developed a collaborative multi-robot system where aerial drones perform target search and communicate GPS coordinates to ground robots. Implemented EKF-based sensor fusion for localization using GPS, IMU, and odometry. Designed a Dynamic Window Approach (DWA) planner with LiDAR-based obstacle avoidance and real-time mapping in unknown environments.
Python ROS Controls Locomotion
Computer Vision
Path Planning