Exploiting the Robot Kinematic Redundancy for Emotion Conveyance to Humans as a Lower Priority Task

Abstract

Current approaches do not allow robots to execute a task and simultaneously convey emotions to users using their body motions. This paper explores the capabilities of the Jacobian null space of a humanoid robot to convey emotions. A task priority formulation has been implemented in a Pepper robot which allows the specification of a primary task (waving gesture, transportation of an object, etc.) and exploits the kinematic redundancy of the robot to convey emotions to humans as a lower priority task. The emotions, defined by Mehrabian as points in the Pleasure -Arousal -Dominance space, generate intermediate motion features (jerkiness, activity and gaze) that carry the emotional information. A map from this features to the joints of the robot is presented. A user study has been conducted in which emotional motions have been shown to 30 participants. The results show that happiness and sadness are very well conveyed to the user, calm is moderately well conveyed, and fear is not well conveyed. An analysis on the dependencies between the motion features and the emotions perceived by This work has been partially supported by the Spanish MINECO projects DPI2011-22471, DPI2013-40882-P and DPI2014-57757-R, the Spanish predoctoral grant BES-2012-054899, and the Japanese challenging exploratory research grant 15K12124. The authors would like to thank the members of the GVLab for their invaluable support with the translations and attending the local participants during the user study.Josep-Arnau Claret Institute of Industrial and Control Engineering, Technical University of Catalonia (UPC), BarcelonaTECH, Barcelona, Catalonia, Spain E-mail: josep.arnau.claret@upc.edu Gentiane Venture Tokyo University of Agriculture and Technology, Tokyo, Japan E-mail: venture@cc.tuat.ac.jp Luis Basañez Institute of Industrial and Control Engineering, Technical University of Catalonia (UPC), BarcelonaTECH, Barcelona, Catalonia, Spain E-mail: luis.basanez@upc.edu the participants shows that activity correlates positively with arousal, jerkiness is not perceived by the user, and gaze conveys dominance when activity is low. The results indicate a strong influence of the most energetic motions of the emotional task and point out new directions for further research. Overall, the results show that the null space approach can be regarded as a promising mean to convey emotions as a lower priority task.

Robot Expressive Motions

“…In Claret et al [15], the authors proposed to generate feasible affective behaviors via null-space decomposition, with the affective components being a lower-priority task compared to the reference task.…”

“…These evaluation methods require extensive experimental setup, time, and there may be variations in each repetition of the experiment due to the system hardware. However, they offer an evaluation closer to the target application and environment, and are more likely to provide an accurate evaluation of the system [15,21,38].…”

“…They provide standardized metrics to assess chiefly affective behaviors. These tests are subjective evaluations and must be processed with statistical analysis with a sufficient number of observers [15,84]. Few studies have compared the validity of the different tests; for example, Saerbeck et al [80] evaluated the impression of the affective behavior of two robots-iCat and Roomba-using both the PANAS and the SAM.…”

Adaptive Fuzzy and Predictive Controllers for Expressive Robot Arm Movement during Human and Environment Interaction

“…PAD emotional model [10] The PAD model has been widely used in studies on virtual agents and computer gaming [12], [13] as well as robotics. A relevant example is described in [14], where the PAD model is mapped in motion features (jerkiness, activity and gaze) to generate affective movement of an anthropomorphic robot. Another approach is shown in [15], where the PAD model and fuzzy logic were applied in a humanoid robot to simulate internal emotional states and reach emotional coherence over time.…”

“…To keep external inputs within temporal restrictions, the optimization is configured using constraints inside of a temporal template Ф(Q) ijoint for each robot joint with maximum amplitude = [10][11][12][13][14][15][16][17][18][19][20], minimum amplitude = [−5, −30], time = [2s-5s], and order of Youla parameters = . This order allows linear parametrization over each of the robot model's transfer functions.…”