Papers
7
Total Citations
1,209
H-Index
7
About
Pat Marion is a leading researcher in humanoid robotics, whose work bridges perception, planning, and human-robot interaction. He is best known for his contributions to the DARPA Robotics Challenge, where he co-developed the optimization-based locomotion, estimation, and control system for the Atlas humanoid robot—a landmark paper with over 814 citations that enabled reliable walking and balancing in complex environments. Marion also pioneered affordance-based perception and whole-body planning, allowing robots to reason about their environment and execute tasks with shared autonomy. His work on the Director user interface (cited over 50 times) advanced teleoperation by giving operators intuitive control over high-degree-of-freedom robots in field scenarios. To improve robotic perception, he developed LabelFusion, a pipeline for generating ground-truth labels from real RGBD data (over 120 citations), which has become a key resource for training deep neural networks in cluttered scenes. Marion’s research has been instrumental in making humanoid robots more capable, autonomous, and practical for real-world deployment.
Research Focus
Key Achievements
Top Papers
- 1
- 2An Architecture for Online Affordance‐based Perception and Whole‐body Planning140 citations · 2014
- 3
- 4Continuous humanoid locomotion over uneven terrain using stereo fusion75 citations · 2015
- 5Director: A User Interface Designed for Robot Operation with Shared Autonomy49 citations · 2016
- 6
- 7