Embedding the Concurrent Autonomous Agent into a Humanoid Robot**This work was supported by CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, process number 9174-11-7), CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) and FAPESB (Fundação de Amparo à Pesquisa do Estado da Bahia).

This work presents the use of the Concurrent Autonomous Agent (CAA) in an Aldebaran NAO humanoid robot. The NAO is a 25 degrees of freedom humanoid with a powerful programmable hardware. The cognitive model of the CAA is composed by three levels the runs concurrently: the cognitive level, the instinctive level and the reactive level. The cognitive level does the planning using a Knowledge Based System (KBS), sending local goals to the instinctive level and receiving from it symbolic information. The instinctive level also uses a KBS, but for navigation purposes, selecting reactive behaviours in the reactive level and receiving from the later processed world perception. The reactive level uses the Unified Humanoid Robotic Software Platform. A rule base is loaded in the instinctive level and experiments with NAO are realized, showing that the robot performed correctly under the CAA.