This article presents the upper-torso design issue of Affeto who can physically interact with humans, which biases the perception of affinity beyond the uncanny valley effect. First, we review the effect and hypothesize that the experience of physical interaction with Affetto decreases the effect. Then, the reality of physical existence is argued with existing platforms. Next, the design concept and a very preliminary experiment are shown. Finally, future issues are given. I. THE UNCANNY VALLEY REVISITED The term “Uncanny” is a translation of Freud’s term “Der Unheimliche” and applied to a phenomenon noted by Masahiro Mori who mentioned that the presence of movement steepens the slopes of the uncanny valley (Figure 2 in [1]). Several studies on this effect can be summarised as follows1. 1) Multimodal impressions such as visual appearance, body motion, sounds (speech and others), and tactile sensation should be congruent to decrease the valley steepness. 2) Antipathetic expressions may exaggerate the valley effect. The current technologies enable us to minimize the gap caused by mismatch among cross-modal factors. Therefore, the valley effect is expected to be reduced gradually. For example, facial expressions and tactile sensations of Affetto [2] are realistic and congruent due to baby-like face skin mask of urethane elastomer gel (See Figure 1). Generated facial expressions almost conquered the uncanny valley. Further, baby-like facial expressions may contribute to the reduction of the valley effect due to 2). In addition to these, we suppose that the motor experience of physical interactions with robots biases the perception of affinity as motor experiences biases the perception of movements [3]. To verify this hypothesis, Affetto needs its body which realizes physical interactions naturally. The rest of this article is organized as follows. The next section argues about the reality of physical existence with existing platforms. Then, the design concept and a very preliminary experiment are shown, and the future issues are given.
[1]
Iseult A. M. Beets,et al.
Nonvisual motor learning improves visual motion perception: evidence from violating the two-thirds power law.
,
2010,
Journal of neurophysiology.
[2]
Junichi Hoshino,et al.
Baby type robot "YOTARO"
,
2009,
SIGGRAPH '09.
[3]
G. Sandini,et al.
The iCub cognitive architecture: Interactive development in a humanoid robot
,
2007,
2007 IEEE 6th International Conference on Development and Learning.
[4]
Chrystopher L. Nehaniv,et al.
Title of paper : KASPAR – A Minimally Expressive Humanoid Robot for Human-Robot Interaction Research
,
2009
.
[5]
G. Rizzolatti,et al.
The mirror-neuron system.
,
2004,
Annual review of neuroscience.
[6]
Masayuki Inaba,et al.
Development of functional whole-body flesh with distributed three-axis force sensors to enable close interaction by humanoids
,
2007,
2007 IEEE/RSJ International Conference on Intelligent Robots and Systems.
[7]
Masaki Ogino,et al.
Cognitive Developmental Robotics: A Survey
,
2009,
IEEE Transactions on Autonomous Mental Development.
[8]
Hiroshi Ishiguro,et al.
Android science: Toward a new cross-interdisciplinary framework
,
2005
.
[9]
Heloir,et al.
The Uncanny Valley
,
2019,
The Animation Studies Reader.
[10]
Kenichi Narioka,et al.
Pneumatic Musculoskeletal Infant Robots
,
2009
.
[11]
Minoru Asada,et al.
Realistic child robot “Affetto” for understanding the caregiver-child attachment relationship that guides the child development
,
2011,
2011 IEEE International Conference on Development and Learning (ICDL).