Guillermo A. Castillo
The Ohio State University, Southern University of Science and Technology
Papers
13
Total Citations
213
H-Index
8
About
Guillermo A. Castillo is a robotics researcher whose work sits at the intersection of reinforcement learning, control theory, and bipedal locomotion. His research focuses on developing robust, hierarchical control frameworks that enable legged robots to navigate complex and unpredictable environments, with a particular emphasis on bridging the gap between data-driven learning methods and classical control techniques. Castillo's most influential contribution — garnering 67 citations — introduced a cascade-structure controller combining reinforcement learning with intuitive feedback regulation for Agility Robotics' Digit platform, demonstrating that principled hybrid architectures can outperform purely end-to-end learning approaches. Subsequent work challenged prevailing assumptions in the field by showing that surprisingly simple linear policies are sufficient to achieve robust walking on uneven and sloped terrains, earning nearly 30 citations each. His earlier exploration merging Hybrid Zero Dynamics with reinforcement learning on the RABBIT robot helped establish a foundational methodology that continues to influence bipedal control research. Beyond locomotion, Castillo has made meaningful contributions to robot safety, developing scenario-based safety testing frameworks and Control Barrier Function-based path planning for obstacle avoidance. His cumulative body of work reflects a sophisticated understanding of both theoretical rigor and real-world deployment challenges, making him a notable emerging voice in legged robotics and safe autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4Reinforcement Learning Meets Hybrid Zero Dynamics: A Case Study for RABBIT21 citations · 2019
- 5Template Model Inspired Task Space Learning for Robust Bipedal Locomotion20 citations · 2023
- 6
- 7
- 8
- 9
- 10