Papers
14
Total Citations
332
H-Index
9
About
Andrei Haidu is a leading researcher in cognitive robotics, specializing in knowledge representation, reasoning, and human-robot interaction. His most significant contribution is the development of KnowRob 2.0, a second-generation knowledge processing framework for cognition-enabled robotic agents, which has garnered 198 citations and stands as a cornerstone in the field. Haidu’s work bridges the gap between abstract human instructions and actionable robot behaviors, notably through automated knowledge acquisition from virtual demonstrations and interactive physics-based simulations. His research on action recognition, failure prediction, and task model acquisition has enabled robots to learn complex manipulation tasks—such as flipping pancakes—by observing human demonstrations in game-like environments. With over 300 total citations across his top papers, Haidu has advanced open robotics research by integrating web-based knowledge services and cloud computation. His notable achievements include developing systems for mixed human-robot rescue teams and action generalization from simple language instructions, pushing the boundaries of how robots acquire common sense and adapt to dynamic scenarios. Haidu’s work is essential reading for anyone interested in making autonomous robots truly intelligent and capable in real-world settings.
Research Focus
Key Achievements
Top Papers
- 1
- 2Open robotics research using web-based knowledge services18 citations · 2016
- 3Cognition-enabled Framework for Mixed Human-Robot Rescue Teams18 citations · 2018
- 4Action recognition and interpretation from virtual demonstrations16 citations · 2016
- 5
- 6Acquiring task models for imitation learning through games with a purpose12 citations · 2013
- 7Learning action failure models from interactive physics-based simulations12 citations · 2015
- 8
- 9Robots, Pancakes, and Computer Games9 citations · 2015
- 10