Bhaskara Marthi
Papers
12
Total Citations
802
H-Index
7
About
Bhaskara Marthi is a leading researcher in robotics and artificial intelligence, whose work bridges the gap between high-level task planning and low-level physical execution. His major contributions lie in combined task and motion planning for mobile manipulation, a field in which his 2010 paper has garnered over 206 citations. Marthi pioneered hierarchical planning systems that enable robots to find high-quality kinematic solutions to complex tasks, such as autonomously navigating offices, opening doors, and plugging into electrical outlets for recharging—a feat demonstrated with remarkable reliability in his 2010 work (176 citations). He has also advanced robot learning from demonstration, developing incremental, semantically grounded methods (193 citations) that allow robots to learn reusable task primitives from unstructured human demonstrations without strong sequencing assumptions. Additionally, Marthi has contributed to semantic world modeling, enabling long-term change detection and object-based reasoning for mobile robots. His research has profound implications for personal robotics, warehouse automation, and assistive technologies, making him a pivotal figure in creating robots that can adapt and operate intelligently in human environments.
Research Focus
Key Achievements
Top Papers
- 1Combined Task and Motion Planning for Mobile Manipulation206 citations · 2010
- 2Learning grounded finite-state representations from unstructured demonstrations193 citations · 2014
- 3Autonomous door opening and plugging in with a personal robot176 citations · 2010
- 4Incremental Semantically Grounded Learning from Demonstration96 citations · 2013
- 5
- 6
- 7Object disappearance for object discovery19 citations · 2012
- 8Sparse Kernel-SARSA(λ) with an Eligibility Trace6 citations · 2011
- 9Unsupervised discovery of object classes with a mobile robot6 citations · 2014
- 10