It is desirable that mobile robots applied to real world applications perform their tasks in previously unknown environments. Thus, a mobile robot architecture capable of adaptation is very suitable. This work presents a hybrid adaptive architecture for mobile robots called AAREACT that has the ability of learning how to coordinate primitive behaviors codified by the Potential Fields method by using Reinforcement Learning. The proposed architecture is evaluated in terms of its performance curve when the robot is moved from a scenario to another. Experiments were performed on a Pioneer robot simulator, from Activ- Media Robotics R . Results suggest that AAREACT has good adaptation skills for specific environment and task