Hao-Ping Hsu
Papers
2
Total Citations
29
H-Index
2
About
Hao-Ping Hsu is a leading researcher in the intersection of reinforcement learning, fuzzy systems, and humanoid robotics. His primary contributions lie in developing novel control and cognition algorithms that enable humanoid robots to perform complex, dynamic tasks. Hsu’s most influential work, "Fuzzy Double Deep Q-Network-Based Gait Pattern Controller for Humanoid Robots" (2020), has garnered 23 citations. In this paper, he pioneered the fusion of adaptive-network-based fuzzy inference systems (ANFIS) with double deep Q-networks (DDQN), creating a Fuzzy DDQN (FDDQN) that allows humanoid robots to generate stable, adaptive gait patterns. This hybrid approach significantly advances robot locomotion by combining the interpretability of fuzzy logic with the learning power of deep reinforcement learning. In his earlier work, "Deep Belief Network–Based Learning Algorithm for Humanoid Robot in a Pitching Game" (2019), Hsu innovatively integrated psychological concepts from Daniel Kahneman’s *Thinking, Fast and Slow*—specifically the dual-process theory of System 1 and System 2—into a deep belief network optimized by inertia weight Particle Swarm Optimization. This work demonstrates his unique ability to cross-pollinate cognitive science with robotics, creating more intuitive and human-like learning algorithms. Hsu’s research is pivotal for the future of autonomous, adaptive humanoid robots.
Research Focus
Key Achievements
Top Papers
- 1
- 2