Dot-to-Dot: Explainable Hierarchical Reinforcement Learning for Robotic Manipulation | IEEE Conference Publication | IEEE Xplore