Chris Atkeson

Papers

Total Citations

161

H-Index

About

Chris Atkeson is a pioneering figure in robotics and machine learning, best known for his foundational work in model-based optimization, approximate dynamic programming, and the simulation-to-reality (sim2real) transfer. His research focuses on enabling robots to learn and perform reliably in complex, real-world environments by bridging the gap between simulated training and physical deployment. Atkeson’s most cited paper, "Sim2Real in Robotics and Automation: Applications and Challenges" (2021, 151 citations), critically examines how large-scale automation can leverage simulation for robust, sustained performance, offering key insights for practitioners and AI supervision. He also advanced approximate dynamic programming with "Random Sampling of States in Dynamic Programming" (2008), introducing sparse sampling and local trajectory optimizers to improve global policy and value function learning. As a spotlight author for "Model-Based Optimization for Robotics" (2014), Atkeson has shaped how researchers design controllers that adapt to real-world physics. His work has profoundly influenced robot learning, manipulation, and human-robot interaction, earning him recognition as a leading voice in making robots more autonomous, data-efficient, and reliable in unstructured settings.