Max Robotics — Global Service Robot Directory + US Certification

About

Yunpeng Pan is a researcher working at the intersection of machine learning, optimization, and autonomous systems, with particular focus on model predictive control (MPC), stochastic optimal control, and reinforcement learning for robotics. His most cited work, "A Neurodynamic Optimization Approach to Nonlinear Model Predictive Control" (2010, 21 citations), demonstrates his early contributions to bridging neural computation and control theory — reformulating complex nonlinear MPC problems as quadratic programming tasks solved via recurrent neural networks, offering a computationally tractable alternative to traditional methods. Pan's subsequent research reflects a consistent drive to make autonomous systems more robust under uncertainty. His work on Bayesian nonparametric approaches to path integral stochastic control (2014) and adaptive probabilistic trajectory optimization via approximate inference (2016) addresses fundamental challenges in sample efficiency and uncertainty quantification in reinforcement learning for physical robotic systems. His 2017 paper on pseudospectral MPC under partially learned dynamics further exemplifies his interest in combining model-based and data-driven techniques to handle incomplete system knowledge. Together, these contributions position Pan as a thoughtful researcher advancing principled, learning-augmented control methods with meaningful real-world applications in robotics and autonomous systems.