The Companion Dog

This was my master thesis in Delft. The idea was to design a flexible architecture to model the behaviour of an emotional companion robotic dog. We deployed the system on AIBO, which is a dog-like autonomous robot, developed by Sony.

The model of personality is an extension of the nPME model, where now needs play the most important role, driving AIBO for every situation. Personality, Mood and Emotions are three more layers that in combination with the needs make AIBO show a dog-like behavior. AIBO will act in an unpredictable and changing environment, and the existing models of personality need improvements, modifications and adaptations to the current situation. So, in the design of the system, the importance of incoming events has been considered.

A modification in this architecture called amygdale approach has been introduced to speed up the resolution of some situations. Besides, the implemented architecture is also designed to be suitable for future extensions in the model. The prototype has been tested in different scenarios using AIBO robot and an extensive user study has been done with several students and workers to test the realism, emotional responses and event coherency in robot’s behavior.

This project focuses on the idea of improving interaction between man and machine through the creation of a biological inspired model of human mind. This artificial brain must act coherently with the environment, drawing conclusions according to the different situations that can arise during a day. To achieve this, it is necessary to create a personality model capable of reasoning about a certain situation to perform complex actions in every context. So there must be a whole coherence between events, actions, and emotions. The situation awareness process would include a reasoning system that combining the different personality parameters (PME) with the levels of certain needs (n) and the incoming events should be able to perform a set of actions to satisfy its needs, in a similar way a real dog would do.

nPME model combines Personality, Mood, Emotions an also ‘needs’ to establish a complex behaviour model. The nPME model we started with a layered PME architecture where mood is seen as an intermediate layer between personality and emotions and therefore is influenced by both. Needs are used to let AIBO act independently in a unpredictable and changing environment (nPME). The inclusion of the needs modifies the role of the mood, and now mood is not directly influenced by emotions or personality, but indirectly as a consequence of the evaluation of the goals, preferences and standards in regards with the events that occur in the environment.

To create a robot autonomous that acts coherently in a unpredictable and changing environment, a main force that drives its behavior must be designed: this force is the satisfaction of the needs. Moreover, the need has to act in combination with the events that defines the environment and always together with the emotional response that defines its attitude towards the situation. Those emotional responses are directly related to the election of the personality and the level of mood. With all this information the system is able to choose a set of actions to perform by AIBO. According to the user study, the approach followed gave very positive results when comparing our robotic dog AIBO with the ideal behavior of a real companion dog.

Special thanks to Zhenke Yang and Leon Rothkrantz from TU Delft for their continuous support and great ideas.

J.Blanco Calvo, D.Datcu, L.J.M.Rothkrantz, AIBO as a needs-based companion dog, Game-On 2007, ISBN 9789077381373, pp. 152-156, November 2007.