Ego-Pi: Egocentric Human Data for Humanoid Manipulation

Ego-Pi explores fine-tuning large Vision-Language-Action (VLA) models for dexterous humanoid manipulation using aligned egocentric human demonstrations and robot data. The project focuses on cross-embodiment generalization, bimanual coordination, and scalable data curation for real-world humanoid learning.

Vision-Language-Action · Humanoid Manipulation · Imitation Learning · Foundation Models

GRaD-Nav-Onboard: Vision-Based Drone Navigation

This project studies onboard, vision-only navigation for drones in dynamic environments. We integrate Gaussian Splatting, velocity estimation networks, and reinforcement learning to enable safe, real-time navigation with dynamic obstacles on embedded platforms.

Autonomous Drones · Reinforcement Learning · Safe Control · Onboard Perception

Safety-Critical Control for Quadruped Robots

This line of work focuses on safety-critical control for quadruped robots using Model Predictive Control (MPC) combined with Control Barrier Functions (CBFs). The approach ensures safe navigation and motion generation under physical and environmental constraints.

Quadruped Robots · MPC · Control Barrier Functions · Safe Robotics