Social Robotics is a research field aiming at designing robots able to interact with people in a natural manner. Within the domain of Socially Assistive Robotics the capability of adapting and personalizing behaviors and assistive services of robots, according to the specific assistive context and needs of a person is crucial to improve the efficacy in users support and hence acceptance. The authors rely on some recent results concerning the realization of a cognitive control approach for assistive robots supporting the synthesis of personalized and flexible assistive behaviors. This paper takes into account a general rehabilitation scenario and presents some initial steps toward the integration of perception, knowledge representation and planning capabilities to pursue flexibility, adaptation and personalization of assistive robot behaviors.
