Papers
7
Total Citations
98
H-Index
4
About
Liz Murphy’s research lies at the intersection of autonomous navigation, topological mapping, and human-aware robotics, with a focus on enabling robots to understand and move through complex, dynamic environments. Her most influential work, “Incremental Unsupervised Topological Place Discovery” (29 citations), introduced an online system that fuses sensory information over time to identify topologically distinct places, challenging the assumption that a single image defines a location. This contribution is foundational for long-term robot autonomy. In earlier work, she developed a general exploration framework using the Gap Navigation Tree to guide SLAM processes (24 citations), and pioneered probabilistic costmaps that explicitly model environmental uncertainty—a departure from traditional assumptive costmaps—demonstrating how these can be learned from vehicle experience (18 citations). Her research also extends to human-robot interaction, as seen in the STALKERBOT project, which enables robots to learn navigational patterns by following people in dynamic human environments. With a career spanning topological exploration, risk-bounded path planning, and traversability mapping, Murphy’s work has shaped how robots build and use spatial representations for safe, efficient, and socially aware navigation.
Research Focus
Key Achievements
Top Papers
- 1Incremental unsupervised topological place discovery29 citations · 2014
- 2
- 3Building Large Scale Traversability Maps Using Vehicle Experience19 citations · 2013
- 4Creating and using probabilistic costmaps from vehicle experience18 citations · 2012
- 5Experimental Comparison of Odometry Approaches4 citations · 2013
- 6Choosing landmarks for risky planning2 citations · 2011
- 7