Joshua A. Haustein
KTH Royal Institute of Technology, Karlsruhe Institute of Technology
Papers
9
Total Citations
334
H-Index
7
About
Joshua A. Haustein’s research lies at the intersection of robotic manipulation, motion planning, and artificial intelligence, with a focus on enabling robots to interact intelligently with cluttered, unstructured environments. His major contributions center on task-specific grasping and non-prehensile rearrangement—that is, moving objects without picking them up, by pushing, sliding, or sweeping. In his highly cited 2017 work on affordance detection (96 citations), Haustein pioneered the use of deep learning to map objects’ task-relevant properties to optimal grasps, bridging perception and action. He also advanced kinodynamic planning for rearrangement, embedding physics models to allow robots to reason dynamically about object interactions, as seen in his 2015 papers (69 and 67 citations). More recently, his application of Monte Carlo tree search to planar non-prehensile sorting (51 citations) has opened new avenues for efficient, real-time multi-object organization. Haustein’s work has been recognized for its practical impact on warehouse automation and domestic robotics, and his hierarchical fingertip grasp planning algorithm (2017) remains a benchmark for dexterous manipulation. With over 330 total citations, his research continues to shape how robots perceive, plan, and act in complex physical spaces.
Research Focus
Key Achievements
Top Papers
- 1Affordance detection for task-specific grasping using deep learning96 citations · 2017
- 2Nonprehensile whole arm rearrangement planning on physics manifolds69 citations · 2015
- 3
- 4
- 5Integrating motion and hierarchical fingertip grasp planning21 citations · 2017
- 6
- 7
- 8
- 9Object Placement Planning and optimization for Robot Manipulators3 citations · 2019