Mark Steedman

University of Edinburgh

Papers

11

Total Citations

324

H-Index

8

About

Mark Steedman is a prominent researcher whose work bridges the critical gap between high-level artificial intelligence planning and low-level robotic control, with significant contributions to cognitive robotics, knowledge representation, and computational linguistics. His most influential contribution, the Object-Action Complex (OAC) framework — introduced in collaborative work that has garnered over 150 citations — provides a principled formalism for grounding symbolic representations in sensorimotor processes, enabling robots to reason about and execute actions in real-world environments. This foundational concept underpins much of his subsequent research, including efforts to integrate cognitive vision systems, automated planning, and execution monitoring on humanoid robot platforms. Steedman has also advanced the field by exploring how common-sense knowledge can be extracted from text to support autonomous robot planning, and by developing machine learning approaches — such as kernel perceptron models — to help robots acquire action knowledge from experience. His 2018 ACL Lifetime Achievement Award, celebrated in "The Lost Combinator," underscores his broader impact on computational linguistics. Across his career, Steedman has consistently tackled the representational discontinuities that challenge truly autonomous, intelligent robotic systems.

Research Focus

Key Achievements

8
H-Index
11
Papers
324
Total Citations
29
Avg Citations/Paper
🏆 Most Cited Paper
Object–Action Complexes: Grounded abstractions of sensory–motor processes
153 citations · 2011
📈 Most Prolific Year: 2014 (2 Papers)
🤝 Key Collaborators: 23
🏛 Institutions: University of Edinburgh

Top Papers

  1. 1
  2. 2
    Object Action Complexes as an Interface for Planning and Robot Control
    66 citations · 2006
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10
    The Lost Combinator
    2 citations · 2018

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago