Yiwen Yang

About

Yiwen Yang is a leading researcher in computational intelligence and robotics, with a primary focus on neural dynamics for real-time optimization and control. Yang’s most influential work centers on Zhang neural networks (ZNN), particularly their application to solving time-varying matrix problems—such as the Moore–Penrose inverse—and quadratic programming under dynamic constraints. A landmark 2010 paper on ZNN for full-rank matrix inversion has garnered 142 citations, establishing a foundational method for handling time-varying systems. Yang further advanced the field by demonstrating the superior robustness of power-sum activation functions in ZNNs, enabling reliable performance even under large implementation errors (48 citations). In a notable 2009 study, Yang revealed the equivalent relationship between velocity- and acceleration-level redundancy-resolution schemes for multi-link robot arms, bridging a gap in decades of separate research. With over 270 total citations across key publications, Yang’s work has significantly shaped the theory and application of neural networks for online optimization, offering robust, real-time solutions critical for robotics and control systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Zhang neural network solving for time-varying full-rank matrix Moore–Penrose inverse

   142 citations · 2010
2. 2
   Performance analysis of gradient neural network exploited for online time-varying quadratic minimization and equality-constrained quadratic programming

   76 citations · 2011
3. 3
   Superior robustness of power-sum activation functions in Zhang neural networks for time-varying quadratic programs perturbed with large implementation errors

   48 citations · 2011
4. 4
   Equivalent relationship between velocity- and acceleration-level redundancy-resolution schemes exemplified via multi-link planar robot arms

   4 citations · 2009

Key Collaborators

Contact & Links