Papers
12
Total Citations
271
H-Index
7
About
Brendan Tidd is a leading robotics researcher whose work sits at the intersection of computer vision, multi-robot systems, and deep reinforcement learning. He is best known for his pivotal contributions to Team CSIRO Data61’s success in the DARPA Subterranean Challenge, where heterogeneous ground and aerial robots explored dangerous, unstructured underground environments. His research on unified perception and autonomy for heterogeneous teams—detailed in papers with nearly 100 citations—demonstrated how balancing robot autonomy with human interaction can achieve robust, adaptive exploration in tunnels and urban infrastructure. Tidd has also advanced human-robot collaboration through a comprehensive survey on robotic vision for interaction (113 citations), and pioneered novel reinforcement learning approaches for robots navigating narrow gaps and complex terrain, including curriculum learning for bipedal walking and residual skill policies for adaptable action spaces. His multi-modal user interfaces for multi-robot control further showcase his commitment to practical, deployable systems. With over 250 total citations and a string of high-impact publications, Tidd is shaping the future of autonomous robotics in challenging, real-world environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5Passing Through Narrow Gaps with Deep Reinforcement Learning8 citations · 2021
- 6
- 7Guided Curriculum Learning for Walking Over Complex Terrain8 citations · 2020
- 8
- 9
- 10