Spencer M. Richards

Vaughn College of Aeronautics and Technology

Papers

Total Citations

107

H-Index

About

Spencer M. Richards is a researcher at the intersection of control theory, robotics, and machine learning, focused on making learning-enabled autonomous systems both adaptive and certifiably safe. His foundational work introduces the Lyapunov Neural Network (2018, 69 citations), a framework that leverages Lyapunov stability theory to provide formal safety guarantees for robots learning in real time—a critical step toward deploying learning algorithms on safety-critical hardware. Richards further advances adaptive control through his work on control-oriented meta-learning (2023, 23 citations), which enables robots to rapidly adapt to uncertain, dynamic environments without sacrificing performance. He also develops contraction-based regularization for learning stabilizable nonlinear dynamics (2020, 8 citations), offering a principled way to ensure robust trajectory tracking. Addressing the practical challenge of reliability, his system-level analysis of out-of-distribution data in robotics (2022, 7 citations) highlights pathways toward trustworthy autonomy. Collectively, Richards’ contributions bridge rigorous control-theoretic guarantees with data-driven learning, empowering robots to operate safely and effectively in the unpredictable real world.