Cheng Sheng

Papers

Total Citations

H-Index

About

Cheng Sheng is a rising researcher in the field of robotics and control systems, with a focus on the intersection of machine learning and autonomous navigation. His primary research areas include auto-tuning of robotic controllers, differentiable programming, and nonlinear dynamics. Sheng’s most notable contribution is the development of **DiffTune**, a groundbreaking framework that leverages auto-differentiation to automatically fine-tune low-level robot controllers. This work addresses a critical challenge in robotics: the manual tuning of controllers for complex, nonlinear systems is notoriously difficult and time-consuming. By enabling gradient-based optimization of controller parameters, DiffTune significantly improves task performance without human intervention. Since its publication in 2022, the paper has already garnered 7 citations, reflecting its growing influence in the robotics community. Sheng’s work is particularly impactful for students and researchers interested in bridging the gap between classical control theory and modern deep learning, offering a practical tool for enhancing robot autonomy. His innovative approach promises to streamline the development of more adaptive and efficient robotic systems.