Papers
66
Total Citations
2,643
H-Index
26
About
George Konidaris is a prominent robotics and artificial intelligence researcher whose work spans robot learning, skill acquisition, symbolic planning, and human-robot interaction. He is perhaps best known for his foundational contributions to the problem of hierarchical skill learning — most notably the Constructive Skill Trees (CST) algorithm, which enables robots to autonomously segment and learn reusable skills from demonstration trajectories, accumulating over 300 citations. His 2018 work bridging low-level motor skills and high-level symbolic representations for planning has proven particularly influential, garnering 244 citations and addressing one of the deepest challenges in intelligent robotics: how agents can construct abstract, provably sound representations suitable for planning in complex, continuous environments. Konidaris has also made significant contributions to robot manipulation, co-authoring a widely read review of the field that has collectively attracted nearly 260 citations across versions. His research extends into hardware-accelerated motion planning, multi-robot coordination under uncertainty, and mixed-reality interfaces for intuitive robot programming and teleoperation. This breadth reflects a vision of robots that are not only capable learners but also safe, transparent collaborators with human users. His work is essential reading for anyone studying the intersection of machine learning, planning, and practical robotics.
Research Focus
Key Achievements
Top Papers
- 1Robot learning from demonstration by constructing skill trees300 citations · 2011
- 2
- 3Learning grounded finite-state representations from unstructured demonstrations193 citations · 2014
- 4
- 5
- 6
- 7
- 8Planning for decentralized control of multiple robots under uncertainty88 citations · 2015
- 9Robot Motion Planning on a Chip87 citations · 2016
- 10End-User Robot Programming Using Mixed Reality78 citations · 2019