Learning of shared attention in sociable robotics

Joint attention is a communicative activity that allows communicators to share perceptual experience by attending on a visual object. This communicative activity is conceptualized as a conditional probability over jointly given target candidates and cues. The joint relationship between them yields attentional cueing effects -, i.e., an enhanced response if the two stimuli are consistent and slower, but a deteriorated response if not. With a computational model, a simulation was carried out to show the interesting properties of the model in which an attentional response is determined by the interaction between the target candidates and the cues. The model successfully locates a visual target object guided by an orientation cue from a pointing gesture. The result indicates that joint attention can be considered as a cooperative decision process on a visual object among many objects with a referential cue driven from an interlocutor.

show abstract

“…(13) to generate an attentional output. An attentional window is allocated at the location that contains the highest attentional output.…”

Section: Integrationmentioning

confidence: 99%

“…Current computational models of joint attention in robotics are conceptualized on the basis of these developmental studies [12,13]. For instance, Nagai et al [14] viewed it as an imitational behavior derived from learning a care giver's head pose, and investigated the developmental relationship between imitation and joint attention.…”

Section: Introductionmentioning

confidence: 99%

Robot's joint attention cued by a pointing gesture

Lee

Shin

2013

Proceedings of the 7th International Conference on Ubiquitous Information Management and Communication

View full text Add to dashboard Cite

show abstract

“…The robotic architecture to be extended in the present work was proposed by Policastro et al [19], [20]. It is constituted by mechanisms and structures inspired on Science of Behavior Analysis [3], [5], [16].…”

Section: The Robotic Architecturementioning

confidence: 99%

“…In this figure, arrows indicate the flow of information in the three modules of the architecture. The circles indicate the methods and component structures of the modules [20].…”

Section: The Robotic Architecturementioning

confidence: 99%

Personal identity module using psychoevolutionary emotion theory for social robots

Silva

Reis

Romero

2012

2012 IEEE RO-MAN: The 21st IEEE International Symposium on Robot and Human Interactive Communication

View full text Add to dashboard Cite

Robots that interacts socially in human being environment should be able to understand and act over it using clues from humans; it includes all types of expression: facial, non-facial, verbal and nonverbal affective expressions. These behaviors need a model capable to compute external and internal information to construct autonomous robots. In this paper, we are proposing a basis for modeling some reflexes inherent to human beings that are responsible by personal identity of each individual. The proposed model is constituted by two reflex modules and an emotion module that are based on psychoevolutionary theory. Then, this model has been inserted in our social architecture as an improvement of old motivational module and evaluated in a simulator. The advantage of the proposed approach from others is the possibility of representing the emotion as an internal function which varies over the time with different types of interaction. The experimental results show that this module is capable to emulate adequately human beings basic reflex in a robot during a simple interaction with a caregiver.

show abstract

“…Há também pesquisas a respeito de sistemas que aprendem a seguir o olhar das pessoas quando fita um objeto, ação esta que é chamada de atenção compartilhada (HOFFMAN et al, 2006;POLICASTRO et al, 2009 …”

Section: Interação Humano-robôunclassified

Arquitetura de controle inteligente para interação humano-robô

Alves¹

View full text Add to dashboard Cite

Learning of shared attention in sociable robotics

Cited by 12 publications

References 20 publications

Robot's joint attention cued by a pointing gesture

Robot's joint attention cued by a pointing gesture

Personal identity module using psychoevolutionary emotion theory for social robots

Arquitetura de controle inteligente para interação humano-robô

Contact Info

Product

Resources

About