Rehj Cantrell
Indiana University Bloomington, Indiana University, Human Media
Papers
8
Total Citations
199
H-Index
6
About
Rehj Cantrell is a leading researcher in human-robot interaction (HRI), specializing in making robot communication more natural, robust, and humanlike. Her work centers on task-based dialogue processing, incremental natural language understanding, and the grounding of spoken instructions in visual perception. In her highly cited work "Tell me when and why to do it!" (60 citations), she addresses the critical need for robots in high-stress domains like search and rescue to communicate effectively with human operators. Cantrell pioneered algorithms for learning action verbs through dialogue ("Learning actions from human-robot dialogues," 47 citations) and developed architectures for robust spoken instruction understanding that process language incrementally, enabling robots to provide timely backchannel feedback. Her research on incrementally biasing visual search using natural language input (2013) demonstrates how tight integration of vision and language processing allows robots to resolve spoken references to objects in real time—a key requirement for fluid interaction. With over 200 total citations, Cantrell's contributions are foundational to creating robots that can understand and act upon human language as naturally as another person would, advancing the frontier of collaborative human-robot teams.
Research Focus
Key Achievements
Top Papers
- 1Tell me when and why to do it!60 citations · 2012
- 2Toward Humanlike Task‐Based Dialogue Processing for Human Robot Interaction48 citations · 2011
- 3Learning actions from human-robot dialogues47 citations · 2011
- 4Robust spoken instruction understanding for HRI15 citations · 2010
- 5Robust spoken instruction understanding for HRI13 citations · 2010
- 6Incrementally biasing visual search using natural language input7 citations · 2013
- 7Toward Human-Like Task-based Dialogue Processing for HRI6 citations · 2011
- 8Incremental Referent Grounding with NLP-Biased Visual Search3 citations · 2012