Max Robotics — Global Service Robot Directory + US Certification

About

Andrew Gordon Wilson is a leading researcher at the intersection of machine learning, robotics, and Bayesian nonparametrics, best known for pioneering scalable deep kernel learning and advancing autonomous systems for extreme environments. His most cited work, "Resilient and Modular Subterranean Exploration with a Team of Roving and Flying Robots" (50 citations), demonstrates a comprehensive multi-robot approach to navigating GPS-denied, communication-limited underground spaces—a breakthrough for search-and-rescue and planetary exploration. Wilson’s foundational contributions to kernel methods include "Learning Scalable Deep Kernels with Recurrent Structure" (64 combined citations), where he introduced expressive closed-form kernels that capture sequential dependencies in speech, robotics, and finance, enabling Gaussian processes to scale to large datasets while retaining uncertainty quantification. He has also critically analyzed physics-inspired neural networks, deconstructing the inductive biases of Hamiltonian Neural Networks (14 citations), and extended Bayesian optimization to high-dimensional outputs (9 citations) for scientific discovery. His work on conditioning sparse variational Gaussian processes for online decision-making (5 citations) further bridges probabilistic modeling with real-time control. With a total of over 140 citations across these papers, Wilson’s research is shaping robust, uncertainty-aware AI systems that operate reliably in complex, real-world settings.