Christopher G. Atkeson

About

Christopher G. Atkeson is a pioneering roboticist and machine learning researcher whose work spans robot learning, humanoid control, nonparametric statistics, and human-robot interaction. Perhaps best known for his foundational contributions to **locally weighted learning** — a memory-based approach to function approximation that has amassed over 1,600 citations and become a cornerstone of nonparametric machine learning — Atkeson has consistently bridged theoretical rigor with practical robotics applications. His early work on estimating inertial parameters of robot manipulators (574 citations) established essential tools for modern robot dynamics. Through landmark studies on humanoid robots, including adapting human motion for robotic control and using humanoid platforms to study human behavior, he has advanced our understanding of both machine and biological movement. His robot juggling experiments elegantly demonstrated memory-based learning on dynamic real-world tasks, while his reinforcement learning contributions, including the Parti-game algorithm, tackled the challenge of high-dimensional state spaces. Most recently, his team's optimization-based control for the DARPA Robotics Challenge and innovative tactile sensing research reflect his enduring commitment to capable, physically intelligent robots. With multiple papers exceeding 200 citations, Atkeson's influence across robotics and machine learning remains profound and wide-reaching.

Research Focus

Key Achievements

Top Papers

1. 1
   Locally Weighted Learning

   1,683 citations · 1997
2. 2
   Estimation of Inertial Parameters of Manipulator Loads and Links

   574 citations · 1986
3. 3
   Adapting human motion for the control of a humanoid robot

   318 citations · 2003
4. 4
   Locally Weighted Learning

   302 citations · 1997
5. 5
   Using humanoid robots to study human behavior

   266 citations · 2000
6. 6
   The Parti-game Algorithm for Variable Resolution Reinforcement Learning in Multidimensional State-spaces

   262 citations · 1995
7. 7
   Robot juggling: implementation of memory-based learning

   243 citations · 1994
8. 8
   Scalable Techniques from Nonparametric Statistics for Real Time Robot Learning

   238 citations · 2002
9. 9
   Optimization‐based Full Body Control for the DARPA Robotics Challenge

   229 citations · 2015
10. 10
    Combining finger vision and optical tactile sensing: Reducing and handling errors while cutting vegetables

    224 citations · 2016

Key Collaborators

Contact & Links