This paper presents a fuzzy logic controller (FLC) for the behaviour implementation of Sony legged robots. The AHC (Adaptive Heuristic Critic) reinforcement learning is employed to refine the FLC. The actor part of AHC is a conventional FLC whose parameters of input membership functions are learned by an immediate internal reinforcement signal. This internal reinforcement signal comes from a prediction of the evaluation value of a policy and the external reinforcement signal. The evaluation value of a policy is learned by TD (temporal difference) learning in the critic part that is also represented by a FLC. GA (Genetic Algorithm) is employed for learning internal reinforcement of the actor part by the considering that they are more efficient in search than other trial and error search approaches.