Chengyu Yang

Papers

Total Citations

H-Index

About

Chengyu Yang is a rising star in robotics and control systems, whose work bridges the gap between classical controller design and modern machine learning. His primary research focuses on the intersection of robotic control, nonlinear dynamics, and differentiable programming—an area where he has made transformative contributions. Yang is best known for developing **DiffTune**, a groundbreaking framework that enables automatic tuning of robot controllers through automatic differentiation. This innovation addresses a long-standing challenge in robotics: the tedious, manual fine-tuning of low-level controllers for high-level task performance. By treating the entire control pipeline as a differentiable system, DiffTune allows gradient-based optimization to replace heuristic tuning, dramatically improving efficiency and robustness. Though early in his career, his most-cited work (13 citations in 2024) has already sparked significant interest, demonstrating the method’s potential across diverse robotic platforms. Yang’s achievements signal a paradigm shift toward data-driven, automated controller synthesis, positioning him as a key figure in the next generation of intelligent robotics research.