A Mind-inspired Architecture for Adaptive HRI

Abstract

One of the main challenges of social robots concerns the ability to guarantee robust, contextualized and intelligent behavior capable of supporting continuous and personalized interaction with different users over time. This implies that robot behaviors should consider the specificity of a person (e.g., personality, preferences, assistive needs), the social context as well as the dynamics of the interaction. Ideally, robots should have a "mind" to properly interact in real social environments allowing them to continuously adapt and exhibit engaging behaviors. The authors' long-term research goal is to create an advanced mind-inspired system capable of supporting multiple assistance scenarios fostering personalization of robot's behavior. This article introduces the idea of a dual process-inspired cognitive architecture that integrates two reasoning layers working on different time scales and making decisions over different temporal horizons. The general goal is also to support an empathetic relationship with the user through a multi-modal interaction inclusive of verbal and non-verbal expressions based on the emotional-cognitive profile of the person. The architecture is exemplified on a cognitive stimulation domain where some experiments show personalization capabilities of the approach as well as the joint work of the two layers. In particular, a feasibility assessment shows the customization of robot behaviors and the adaptation of robot interactions to the online detected state of a user. Usability sessions were performed in laboratory settings involving 10 healthy participants to assess the user interaction and the robot's dialogue performance.

Keywords: Assistant Robotics; Engaging and personalized HRI; Mind-inspired architectures for social robots; Multi-modal interaction.

Fig. 1

A sketch of the architecture

Fig. 2

Internal components of System 2 level

Fig. 3

Internal components of the Reactive Reasoner (System 1)

Fig. 4

Ranking of stimulation capabilities for the considered user profiles. The dotted line shows for each profile the computed average ranking values of the exercises

Fig. 5

Detailed view of rankings with respect to the stimuli of Table 2

Fig. 6

Gantt representation of an executed timeline-based plan

Fig. 7

Deployment of Miriam on a Sanbot Elf

Fig. 8

A personalized dialogue sketch showing adaptation to different users’ responses

Fig. 9

A personalized dialogue sketch showing possible answers based on users’ personality

Fig. 10

Interaction length measured in minutes for each participant

Fig. 11

Number of users turns and robot turns per each participant

Fig. 12

Scores obtained at the CUQ by each participant