In the literature, several works are reported about the gait of bipedal robots. Some papers are associated with stability or automatic learning and navigation, but no paper was found that includes the variables of bipedal gait, stability, navigation and automatic learning, all applied in an indoor bipedal robot. Our objective is, after the analysis of the state of the art, to propose a methodology that will allow the autonomous navigation of a bipedal robot and validate its degree of autonomy with respect to the evaluation metrics reported in the literature.