Yuecheng Mao

About

Yuecheng Mao is a researcher at the intersection of robotics and machine learning, specializing in learning from demonstration, movement primitives, and vision-based robotic control. His most influential work, "Learning deep movement primitives using convolutional neural networks" (53 citations), addresses a critical limitation in task-parameterized dynamic movement primitives (TP-DMPs)—their reliance on custom vision systems. Mao pioneered the integration of convolutional neural networks to directly extract task-relevant variables from raw visual input, enabling TP-DMPs to generalize to real-world, unstructured environments without specialized hardware. This contribution bridges deep learning and classical robotics, allowing robots to adapt learned skills to novel situations using only camera data. His work has been cited by researchers advancing robot learning, imitation learning, and adaptive control systems. By removing the need for engineered perception pipelines, Mao’s approach has practical implications for deploying flexible robotic systems in manufacturing, healthcare, and service robotics. His research continues to shape how robots perceive and interact with dynamic environments, making learned movements more robust and accessible for real-world applications.

Research Focus

Key Achievements

Top Papers

1. 1
   Learning deep movement primitives using convolutional neural networks

   53 citations · 2017

Key Collaborators

Contact & Links