Descripción del proyecto
Cómo se llevan los robots con los humanos
Los robots actuales interactúan con las personas al realizar numerosas tareas, lo que suele implicar ciertos problemas, ya que nuestra percepción del mundo depende tanto de la información sensorial que captamos como de nuestros conocimientos previos. Hay que entender mejor la complejidad de estas dinámicas en la interacción social y su influencia en una comprensión eficaz. Con esto en mente, en el proyecto financiado con fondos europeos wHiSPER, se llevará a cabo un estudio en el que se investigará por primera vez cómo cambian los mecanismos básicos de inferencia perceptiva espaciotemporal durante dichas interacciones. Se aplicarán métodos psicofísicos y modelos bayesianos para explicar el papel de los nuevos contextos interactivos entre personas y robots. Los resultados permitirán a se desarrolle tecnología para una colaboración eficaz y adaptable entre los robots y las personas que esté centrada en el cuidado del bienestar de estas últimas.
Objetivo
Perception is a complex process, where prior knowledge is incorporated into the current percept to help the brain cope with sensory uncertainty. A crucial question is how this mechanism changes during interaction, when the brain is faced with two conflicting goals: either optimizing individual perception by using internal priors, or maximizing perceptual alignment with the partner, by limiting the reliance on individual priors. wHiSPER proposes to study for the first time how visual perception of space and time is modified during interaction, by moving the investigation to an interactive shared context, where two agents dynamically influence each other. To allow for scrupulous and systematic control during interaction, wHiSPER will use a humanoid robot as a controllable interactive agent. The research will be articulated along five main objectives: i) determine how being involved in an interactive context influences perceptual inference; ii) assess how perceptual priors generalize to the observation of others actions; iii) understand whether and how individual perception aligns to others priors; iv) assess how is it possible to enable shared perception with a robot and v) determine whether perceptual inference during interaction is modified with aging, when lowered sensory acuity could increase priors relevance. To these aims wHiSPER will exploit rigorous psychophysical methods, Bayesian modeling and human-robot interaction, by adapting well-established paradigms in the study of visual perception to a novel interactive context. In several experiments the humanoid robot and the participants will be shown simple temporal or spatial perceptual stimuli that they will have to perceive either to reproduce them or to perform a coordinated joint action (as passing an object). The measures of the reproduced intervals and of the kinematics of the actions will allow to quantify through Bayesian modeling how social interaction influences visual perception.
Palabras clave
Programa(s)
Régimen de financiación
ERC-STG - Starting GrantInstitución de acogida
16163 Genova
Italia