Yule Cai

I've developed autonomous robot navigation research at Universiti Utara Malaysia, including a decoupled RL-LLM architecture for robust navigation under partial observability. My first-author work evaluated 1,750 episodes across five difficulty levels and seven baselines, achieving 95% mean success among learning/reactive methods and validating edge deployment on an Orange Pi 4.

I've also contributed to kinematic-aware mobile robot path planning, explainable machine learning for aircraft-engine remaining useful-life prediction, and a computer-vision-to-manipulation pipeline for ASL gestures using the Shadow Dexterous Hand. I publish in robotics and AI research and serve as a peer reviewer for Springer Nature journals.