Papers
7
Total Citations
59
H-Index
4
About
Yuka Ariki is a roboticist whose research lies at the intersection of motion planning, imitation learning, and humanoid robotics. Her most impactful work addresses the fundamental challenge of enabling robots to plan collision-free paths efficiently in complex, high-dimensional spaces. Ariki pioneered the integration of 3D convolutional neural networks with sampling-based planners, creating a heuristic-guided framework that dramatically accelerates task-space motion planning—a contribution that has garnered 26 citations and stands as her most recognized work. She further advanced this line of inquiry by developing fully convolutional architectures that learn search heuristics for rapid path planning, directly tackling the real-time constraints that plague robotic manipulation and navigation. Beyond planning, Ariki has made notable contributions to humanoid robotics, where she developed methods for extracting primitive representations from human motion and ground reaction force data to generate physically consistent imitated behaviors. Her work on latent Kullback-Leibler control for dynamic imitation learning enables humanoid robots to replicate whole-body behaviors while maintaining balance—a critical achievement for deploying humanoids in real-world environments. Through her research, Ariki has helped bridge the gap between deep learning and classical robotics, making autonomous systems faster, safer, and more human-like in their movements.
Research Focus
Key Achievements
Top Papers
- 13D-CNN Based Heuristic Guided Task-Space Planner for Faster Motion Planning26 citations · 2020
- 2Fully Convolutional Search Heuristic Learning for Rapid Path Planners11 citations · 2019
- 3
- 4
- 5Observing human movements to construct a humanoid interface3 citations · 2014
- 6
- 7