Papers
5
Total Citations
152
H-Index
3
About
Po-Chien Luan is a leading researcher at the intersection of humanoid robotics and artificial intelligence, specializing in human-robot interaction, emotion recognition, and gait control. His most impactful work, a CNN and LSTM-based facial expression analysis model for humanoid robots (106 citations), enables robots to perceive and respond to human emotions in real time, fundamentally improving the quality of human-robot interaction. Luan has also made significant contributions to bipedal locomotion, developing a fuzzy double deep Q-network-based gait pattern controller (23 citations) and a sequential sensor fusion-based LSTM controller (19 citations) that allow humanoid robots to walk with greater stability and adaptability. His recent work pushes into human motion prediction, with the Multi-Transmotion pre-trained model and a unified monocular vision system for localization and trajectory prediction, addressing the critical challenge of robust, real-world deployment. By combining deep reinforcement learning, fuzzy systems, and sensor fusion, Luan’s research bridges the gap between controlled lab environments and practical robotic applications, making him a key figure in advancing socially intelligent and physically capable robots.
Research Focus
Key Achievements
Top Papers
- 1CNN and LSTM Based Facial Expression Analysis Model for a Humanoid Robot106 citations · 2019
- 2
- 3
- 4
- 5Multi-Transmotion: Pre-trained Model for Human Motion Prediction2 citations · 2024