|
|
|
FJA@ROBOPHILOSOPHY2016 > Home
in the framework of
Aarhus, Denmark,
October 18, 2016
Presentation It is interesting to observe, from a roboticist point of view, that human-human joint action is a topic of intense research in cognitive psychology and philosophy. This observation led us to implement a multi-disciplinary initiative to create a unique opportunity for scientific exchange through a series of workshops called "toward a Framework for Joint Action" (fja.sciencesconf.org). Psychologists and philosophers can present recent developments in joint action research, while roboticists are able to discuss the challenges they face with regard to human-robot interaction and more precisely human-robot joint activity. For this sixth edition, we will focus on Commitment and agency management in joint action. Robots are becoming increasingly flexible agents, able to perform a broader range of actions in diverse contexts, and to adapt their actions to coordinate with human co-actors. This increasing flexibility brings with it new challenges: the more flexible a robot agent is, the more uncertainty a human agent may have about how the robot is going to act. Moreover, robots have to identify and keep track of various action options, and to prioritize some action options. In meeting this challenge, roboticists have been developing various capacities and features that facilitate predictability and communication, from eye gaze and kinematics, to emotional expressions and responsiveness to human emotional expression. In addition, some researchers have worked on higher-level cognitive features, such as the sense of commitment, which may lead robots to resist distraction or alternative action options, and may lead humans also to expect robots to do so, and thus to be willing to rely upon robots. The current workshop will aim towards the articulation of goals and benchmarks toward which this research aims.
Speakers
|
| Online user: 50 |
|
