Kirk MacTavish
Papers
9
Total Citations
229
H-Index
9
About
Kirk MacTavish is a leading researcher in long-term autonomous navigation, specializing in vision-based localization and simultaneous localization and mapping (SLAM) for mobile robots operating in challenging, GPS-denied environments. His most influential work centers on the "Visual Teach and Repeat" (VT&R) paradigm, where robots autonomously retrace manually taught routes over vast distances using only inexpensive cameras. MacTavish’s 2016 paper, "Bridging the appearance gap," with 82 citations, introduced multi-experience localization to overcome environmental changes from lighting and weather, a critical breakthrough for persistent outdoor autonomy. He further advanced this with VT&R 2.0, demonstrated in a 2017 field test covering miles of rugged terrain. Beyond wheeled robots, MacTavish pioneered tethered SLAM (TSLAM) for steep, dangerous terrains, and relative continuous-time SLAM for rolling-shutter cameras. His work on night-time visual odometry using headlights and colour-constant image place recognition showcases his ability to solve real-world perception challenges. With over 200 total citations across his top papers, MacTavish’s contributions have expanded the practical limits of vision-based navigation, enabling robust, long-term autonomy for field robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4TSLAM: Tethered simultaneous localization and mapping for mobile robots22 citations · 2017
- 5Relative continuous-time SLAM18 citations · 2015
- 6Night Rider: Visual Odometry Using Headlights15 citations · 2017
- 7Beyond a Shadow of a Doubt: Place Recognition with Colour-Constant Images12 citations · 2016
- 8
- 9Towards Visual Teach and Repeat for GPS-Denied Flight of a Fixed-Wing UAV10 citations · 2017