Toward the Development of Cognitive Robots
With the aim of endowing robots with the ability to engage people in real social interactions, it is currently typical that novel architectures for robot control take into account in its internal design concepts and schemes originated from cognitive theories. The objective is that robots will be able to emanate responses at human interaction rates and exhibit a pro-active behaviour. This pro-active behaviour implies that the internal architecture of these robots should not only be able to perceive and act. It should also be able to perform off-line reasoning. This paper introduces RoboCog, a new cognitive architecture whose four core elements are a deep -in the concrete-abstract dimension- and hybrid -in the symbolic-numeric dimension- representation of the current state, including the robot itself and the observed world; a set of agents that provide broad functionalities such as navigation, body movement control, dialog or object recognition, and as a result build and maintain this representation; an internal emulation and planning facility where foreseen courses of action can be inferred and tested; and an Executive module that coordinates the interactions among all others. Furthermore, agents themselves can reproduce internally this architecture, including their own replicas of the four elements. The typical scheme of 3-tier architectures is therefore replaced by a recursive estructure that provides a more flexible scenario, where the responses of all agents are tied together by the use of a common inner representation. Preliminar results of the proposed architecture in real scenarios show how RoboCog is able to enhance the effectiveness and time-of-response of complex multi-degree-of-freedom robots designed to collaborate with humans.
KeywordsRobotics Cognitive architectures Simulation theory of cognition
This contribution has been partially granted by the Spanish Government and FEDER funds under coordinated project TIN2012-38079. RoboCog is a cognitive architecture being developed as a common effort among several research groups at different universities including the University of Extremadura (RoboLab), the University of Castilla-La Mancha (SIMD), the University of Málaga (ISIS and GISUM), the University Carlos III of Madrid (PLG) and the University of Jaén (M2P). The robotic therapy for motor rehabilitation is conducted with the group of Technological Innovation (GIT) and the Rehabilitation Department of the Hospital Universitario Virgen del Rocío (HUVR, Seville).
- 1.Alcázar, V., Guzmán, C., Milla, G., Prior, D., Borrajo, C., Castillo, L., Onaindia, E.: PELEA: planning, learning and execution architecture. In: Proceedings of the 28th Workshop of the UK Planning and Scheduling Special Interest Group, Brescia-Italy (2010)Google Scholar
- 2.Bustos, P., Martínez-Gómez, J., García-Varea, I., Rodríguez-Ruiz, L., Bachiller, P., Calderita, L., Manso, L.J., Sánchez, A., Bandera, A., Bandera, J.P.: Multimodal interaction with Loki. In: Workshop on Agentes Físicos, Madrid-Spain (2013)Google Scholar
- 3.Calderita, L.V., Bustos, P., Suárez-Mejías, C., Fernández, F., Bandera, A.: Therapist: towards an autonomous socially interactive robot for motor and neurorehabilitation therapies for children. In: Patients Rehabilitation Research Techniques Workshop (REHAB2013), Venice-Italy (2013)Google Scholar
- 6.Gat, E.: On three-layer architectures. In: Kortenkamp, D., Bonnasso, R.P., Murphy, R. (eds.) Artificial Intelligence and Mobile Robots, pp. 195–210. MIT Press, Cambridge (1998)Google Scholar
- 12.Kawato, M., Wolpert, D.: Internal models for motor control. In: Bock, G.R., Goode, J.A. (eds.) Novartis Foundation Symposium 218: Sensory Guidance of Movement, pp. 291–307. Wiley, Chichester (1998)Google Scholar
- 13.Pacherie, E.: The sense of control and the sense of agency. Psyche 13(1), 1–30 (2007)Google Scholar
- 14.Prinz, W.: Experimental approaches to action. In: Roessler, J., Eilan, N. (eds.) Agency and Self-Awareness, pp. 175–187. Oxford University Press, Oxford (2003)Google Scholar
- 18.Stein, L.A.: Imagination and Situated Cognition. AI Memo 1277, MIT AI Lab (1991)Google Scholar