Max Robotics — Global Service Robot Directory + US Certification

About

Shaofeng Du is a robotics researcher whose work bridges the gap between industrial automation and human-robot interaction, with a particular focus on intelligent perception and skill learning. His most cited work, "Welding Seam Tracking in Robotic Gas Metal Arc Welding" (96 citations), addresses a critical challenge in manufacturing by enabling robots to autonomously follow weld seams, improving precision and reducing human error. Du has also made significant contributions to 3D reconstruction for welding, developing methods to identify initial welding positions from point cloud data (41 citations). Beyond industrial applications, his research extends to wearable robotics, where he studied the effect of a mobile waist assist robot on reducing lower back pain during lifting tasks. In the realm of deep learning, Du has tackled complex perception problems, including picking point detection in dense, occluded environments and imitation learning for precision manipulation tasks like needle reaching. His recent work on variational information bottleneck regularized deep reinforcement learning (2023) aims to make robotic skill adaptation more efficient and safer for real-world deployment. With over 160 citations across his key publications, Du’s research demonstrates a consistent focus on making robots more perceptive, adaptive, and useful in both industrial and assistive contexts.