<FONT face=Verdana>This paper focuses on the recurrence component of imitation mechanism and presen<FONT face=Verdana>ts humanoid reasoning strategies such as knowledge use and action preview to imp<FONT face=Verdana>lement effective behavior imitation.Concretely
by defining the Hausdorff distan<FONT face=Verdana>ce between fuzzy sets as the metric of knowledge use
we combined knowledge radiu<FONT face=Verdana>s into the distancetype fuzzy reasoning method for the implementation of selec<FONT face=Verdana>tive knowledge use;according to the main idea of preview control
we designed the<FONT face=Verdana> action preview model as high level decision of knowledge use to optimize the pa<FONT face=Verdana>rameters of knowledge use strategy.Finally we presented the results of driving b<FONT face=Verdana>ehavior imitation
demonstrated the validity of humanoid reasoning strategies in<FONT face=Verdana>dynamic knowledge use
and realized quick learning and imitating performance.