Michael Erdmann

About

Michael Erdmann is a pioneering roboticist whose work spans motion planning under uncertainty, dexterous manipulation, and action-based sensing. A faculty member at Carnegie Mellon University, Erdmann has spent decades advancing our understanding of how robots can act purposefully in unpredictable, real-world environments. His early landmark contribution, "On Motion Planning with Uncertainty" (1984, 170 citations), laid critical groundwork for how robots navigate errors in modeling, sensing, and control — a foundational challenge still central to robotics today. He extended this thinking through probabilistic and randomized strategies, exploring how deliberate randomization can help robots complete tasks despite incomplete information, as seen in his influential work on probabilistic robot strategies (1989, 38 citations) and randomization in robot tasks (1992, 46 citations). Erdmann has also made significant strides in manipulation robotics. His research on extrinsic dexterity (2014, 279 citations) — his most-cited work — reframed in-hand manipulation by leveraging external forces rather than complex mechanical fingers. His inventive "mobipulator" robot demonstrated that even wheel-driven platforms could perform elegant manipulation tasks. Across more than 800 cumulative citations, Erdmann's career reflects a rare blend of theoretical rigor and imaginative engineering that continues to inspire researchers in robotics and autonomous systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Extrinsic dexterity: In-hand manipulation with external forces

   279 citations · 2014
2. 2
   On Motion Planning with Uncertainty

   170 citations · 1984
3. 3
   An Exploration of Nonprehensile Two-Palm Manipulation

   85 citations · 1998
4. 4
   Understanding Action and Sensing by Designing Action-Based Sensors

   79 citations · 1995
5. 5
   Randomization in Robot Tasks

   46 citations · 1992
6. 6
   Pose and Motion from Contact

   42 citations · 1999
7. 7
   A mobile manipulator

   38 citations · 2003
8. 8
   On probabilistic strategies for robot tasks

   38 citations · 1989
9. 9
   Algorithmic Foundations of Robotics VI

   32 citations · 2005
10. 10
    Randomization for robot tasks: Using dynamic programming in the space of knowledge states

    28 citations · 1993

Key Collaborators

Contact & Links