|
Embodied Affect in Autonomous and Social Robots| old_uid | 12788 |
|---|
| title | Embodied Affect in Autonomous and Social Robots |
|---|
| start_date | 2016/12/07 |
|---|
| schedule | 10h |
|---|
| online | no |
|---|
| summary | Why would we give emotions to robots? Emotions are a key aspect of human behavior, communication and cognition. Affective computing and human-oriented approaches to technology design take into account different aspects of emotions and their expression in order to create technology that humans feel comfortable with. However, the potential uses of emotions in robotics go beyond their social and communicative functions. In this talk, I will discuss, and illustrate with examples of research and systems developed in my group, how emotion modeling in autonomous and social robotics can serve two main purposes: 1) Understanding emotions in biological systems (humans and other animals) and their functions. From this perspective, we investigate questions such as: — What are the key features of human and animal emotions that can be shared by autonomous robots in similar environments? — How do emotions affect behavior and cognition? — How did emotions evolve? — How do they develop through life? 2) Building better robots. From this perspective, we try to answer questions such as: — How can emotions help robots to make ”good” decisions? — How can emotions make robots interact with people better? I will also discuss how, in my view, answering these questions requires giving robots internal models of emotions in addition to skills to express emotions and recognize emotions, and how these models need to be embodied. |
|---|
| responsibles | Baumard |
|---|
| |
|