Max Robotics — Global Service Robot Directory + US Certification

About

Lilai Yan is a researcher specializing in intelligent control systems, recurrent neural networks (RNNs), and machine learning applications in robotics and mechanical engineering. Working primarily in the mid-to-late 1990s, Yan made notable contributions to the development of neural network-based control strategies for nonlinear dynamic systems whose behaviors are initially unknown or difficult to model analytically. Yan's most influential work introduced a sophisticated feedforward-feedback-learning controller architecture that leveraged RNNs to approximate the inverse dynamics of robotic systems, enabling adaptive, model-based control without requiring prior knowledge of plant dynamics. This approach, cited 22 times, represented a meaningful advance in robot learning control. Complementing this, Yan contributed efficient quasi-Newton learning algorithms for training RNNs in mechanical system modeling, demonstrating practical improvements in computational efficiency for nonlinear system identification. Additional work explored direct neural network learning controllers using Powell's optimization method, further broadening the toolkit available for controlling unknown dynamic systems without explicit model identification. Across a focused but impactful body of research, Yan helped establish foundational techniques at the intersection of neural computation and engineering control, contributing ideas that informed subsequent generations of intelligent robotics and adaptive control research.