Junn Yong Loo
Papers
8
Total Citations
160
H-Index
6
About
Junn Yong Loo is a robotics researcher specializing in soft robot sensing, state estimation, and deep learning-based perception systems. His work addresses one of the most fundamental challenges in soft robotics: the difficulty of integrating sensors into compliant, flexible structures without compromising their mechanical properties. To overcome this barrier, Loo has pioneered indirect sensing approaches that leverage robot dynamics, observer theory, and advanced estimation schemes to infer critical variables such as curvature, force, and system states without direct measurement. His most influential contribution, "Robust Multimodal Indirect Sensing for Soft Robots Via Neural Network-Aided Filter-Based Estimation" (2021, 61 citations), demonstrates the power of combining neural networks with filter-based estimation for multimodal perception. Complementing this, his work on predictive uncertainty estimation and H-infinity-based Extended Kalman Filters reflects a sophisticated understanding of probabilistic and robust control frameworks. More recently, Loo has expanded into data-efficient deep learning, exploring synthetic data generation and cross-domain transfer learning to address the scarcity of training data in soft robotic systems. With over 160 cumulative citations, his research meaningfully advances the autonomy and reliability of soft robots destined for unstructured, real-world environments, establishing him as a promising voice in intelligent soft robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5A Deep Learning Framework for Soft Robots with Synthetic Data13 citations · 2023
- 6
- 7
- 8