My research focuses on developing intelligent and reliable robotic systems that can better understand human intentions
and translate them into effective actions in the physical world. In particular, I am interested in combining
trajectory optimization and generative models to build adaptive robotic systems for real-world applications.
This direction has also led me to recently explore onboard humanoid-object interaction, which I view as a coupled problem
involving high-level interactive motion planning and low-level general tracking.
Pulications & Projects
Learning LiDAR-Based Dynamic Obstacle Avoidance via Time-to-Collision Supervision
This project develops a multi-UAV system integrating LiDAR and vision sensing for
autonomous navigation in complex environments while maintaining precise formation control.
Auto filmer: Autonomous aerial videography under human interaction
