Application of Hierarchical Reinforcement Learning in Robotic Soccer
Hong Li · Jisuanji fangzhen · 2005
Robotic soccer is a challenging research domain because many different research areas have to be addressed in order to create a successful team of players such as computer , artificial intelligent , vision and mechanism etc. Player's intelligence can mainly be embodied in learning capacity. One problem in robotic soccer is to adapt skills and the overall behavior in a changing environment. We applied hierarchical reinforcement learning in a SMDP framework learning on all levels simultaneously. As our experiments show, learning simultaneously on the skill level and on the skill selection level is advantageous.