John Welsh
Papers
2
Total Citations
10
H-Index
2
About
John Welsh is a robotics researcher whose work focuses on enabling robots to navigate, understand, and interact with complex, long-horizon environments. His key research areas include spatio-temporal memory for robot navigation, human-robot interaction, and real-time person detection and re-identification. Welsh’s most notable contribution is the development of ReMEmbR, a system that allows robots to build and reason over long-horizon spatio-temporal memory, enabling them to answer questions about where and when events occurred in their environment—a significant leap for autonomous navigation and human-robot collaboration. This work, published in 2025, has already garnered 7 citations, reflecting its early impact. His earlier research on real-time pose-based human detection and re-identification for robot person following, published in 2017, laid the groundwork for robust, single-camera tracking systems. By leveraging convolutional neural networks, Welsh advanced the state of the art in person-following robots, addressing key challenges in dynamic environments. His work is essential reading for students and researchers interested in bridging memory, perception, and action in robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2