Takahiro Miki
Papers
20
Total Citations
2,001
H-Index
16
About
Takahiro Miki is a leading robotics researcher specializing in legged locomotion, terrain perception, and autonomous navigation for quadrupedal and wheeled-legged robots. His work sits at the intersection of machine learning, motion planning, and field robotics, with a particular focus on enabling robots to operate reliably in complex, unstructured real-world environments. Miki's most influential contribution, "Learning Robust Perceptive Locomotion for Quadrupedal Robots in the Wild" (2022, 729 citations), demonstrated how combining exteroceptive perception with learned locomotion policies allows robots to traverse challenging terrain with remarkable agility and efficiency — a landmark result in the field. This work set a new benchmark for autonomous legged systems operating beyond laboratory conditions. His contributions extend to multi-robot exploration, having been part of Team CERBERUS, which won the prestigious DARPA Subterranean Challenge (225 citations), and to planetary analog exploration using teams of legged robots (104 citations). He has also advanced GPU-accelerated elevation mapping, mobile manipulation, and dexterous limb control in quadrupeds. Across his career, Miki has accumulated well over 1,800 citations, reflecting his substantial and growing impact on autonomous robotics research worldwide.
Research Focus
Key Achievements
Top Papers
- 1Learning robust perceptive locomotion for quadrupedal robots in the wild729 citations · 2022
- 2CERBERUS in the DARPA Subterranean Challenge225 citations · 2022
- 3Robust Rough-Terrain Locomotion with a Quadrupedal Robot200 citations · 2018
- 4Perceptive Locomotion in Rough Terrain – Online Foothold Optimization109 citations · 2020
- 5
- 6Elevation Mapping for Locomotion and Navigation using GPU105 citations · 2022
- 7
- 8
- 9
- 10Circus ANYmal: A Quadruped Learning Dexterous Manipulation with Its Limbs52 citations · 2021