How can we sustain the production infrastructure of agriculture as its workforce declines? I study teleoperation that lets a single operator supervise multiple machines, machine learning for the intelligence and safety of agricultural robots, and the field network infrastructure that supports them.
Environmental & Agricultural Science / Agricultural Environmental Engineering, Agricultural Information Engineering
Researching automation for agricultural robots that operate in real fields, including robot tractors and autonomous field vehicles. My work combines teleoperation, autonomous navigation, 3D ToF sensing in orchards, and vehicle control under communication delay to build robotic systems that can work safely and reliably without requiring workers to be physically present in the field.
Research on next-generation network infrastructures and evaluation of network delay impacts for teleoperation of agricultural robots. I focus especially on WebRTC-based real-time communication and the development of low-latency, stable remote monitoring and operation systems for agricultural robotic machinery.
Building safety systems for agricultural robots using CNNs, extracting affordances and controlling agricultural robots with VLMs, and capturing human expertise through imitation learning. By combining image recognition, vision-language models, and teleoperation data, I aim to translate skilled human judgment into robotic intelligence for advanced automation of agricultural work.
Project: A Study on Agricultural Robots Integrating Remote Driving and Reinforcement Learning
Link