Craig Knuth

Johns Hopkins University Applied Physics Laboratory

Papers

Total Citations

H-Index

About

Craig Knuth is an emerging robotics researcher whose work sits at the intersection of autonomous navigation, machine learning, and safety-critical control systems. His research focuses on enabling robots to operate reliably in challenging, real-world environments where traditional methods often fall short. In his most-cited work, "Complex Terrain Navigation via Model Error Prediction" (2022, 14 citations), Knuth tackles a fundamental limitation of conventional geometric navigation approaches — their inability to handle deformable terrain. By leveraging predictive modeling of errors rather than rigid geometric assumptions, his method opens new possibilities for robots navigating soft, dynamic, or unpredictable surfaces. Complementing this, his 2023 paper on statistical safety and robustness guarantees (5 citations) addresses one of the most pressing challenges in modern robotics: providing formal assurances that autonomous systems will remain safe and reach their goals even when operating under unknown, nonlinear stochastic dynamics. By jointly learning dynamics models from data while preserving statistical guarantees, this work bridges the gap between data-driven flexibility and rigorous safety requirements. Together, these contributions mark Knuth as a thoughtful researcher advancing the frontier of dependable autonomous systems in complex, uncertain environments.