Jon Arrizabalaga

Papers

Total Citations

219

H-Index

About

Jon Arrizabalaga is a leading researcher at the intersection of machine learning, model predictive control (MPC), and agile robotics. His primary contributions lie in developing real-time, learning-based control systems that push the boundaries of speed and precision for autonomous platforms. His most influential work, "Real-Time Neural MPC: Deep Learning Model Predictive Control for Quadrotors and Agile Robotic Platforms" (2023, 191 citations), pioneered a framework that integrates deep neural networks directly into the MPC loop, enabling quadrotors to execute aggressive maneuvers while maintaining real-time computational feasibility. This work has become a cornerstone for high-performance autonomous flight. Beyond quadrotors, Arrizabalaga has advanced motion planning for mobile robots by incorporating caster-wheel dynamics into MPC, and has tackled the challenging problem of agile liquid transportation with robotic manipulators, achieving real-time slosh-free tracking. His research on time-optimal tunnel-following for quadrotors further demonstrates his commitment to pushing the limits of constrained, dynamic navigation. With a growing citation impact and a focus on bridging deep learning with real-world control, Arrizabalaga is shaping the future of autonomous systems that must operate at the edge of their physical capabilities.