Natural Tasking of Robots Based on Human Interaction Cues
نویسنده
چکیده
When a person gives a task to another person there are at least two sorts of very human processes at work. At the surface level, each person both displays and perceives cross-cultural cues which regulate the interaction. Through facial expressions, body posture, and utterances, the student unconsciously speeds or slows the rate at which the instructor is teaching and directs the instructor to provide more information when necessary. At a deeper level, the transfer of information is successful because both student and instructor share a common sense of how the world works. Both student and teacher share not only knowledge about how objects behave (an intuitive physics) but also knowledge about how other people behave (an intuitive psychology). Our challenge is to make commanding robots as intuitive and natural as commanding professional soldiers by providing a natural and intuitive interface that capitalizes on a person’s intuitive understanding of how to communicate, and by instilling into robots that same deep understanding of the world which is shared by people.
منابع مشابه
The Essential Dynamics Algorithm: Essential Results
This paper presents a novel algorithm for learning in a class of stochastic Markov decision processes (MDPs) with continuous state and action spaces that trades speed for accuracy. A transform of the stochastic MDP into a deterministic one is presented which captures the essence of the original dynamics, in a sense made precise. In this transformed MDP, the calculation of values is greatly simp...
متن کاملTowards Natural Human-Robot Interaction using Multimodal Cues
Robust human-robot interaction in dynamic domains requires that the robot autonomously learn from sensory cues and adapt to unforeseen changes. However, the uncertainty associated with sensing and actuation on mobile robots makes autonomous operation a formidable challenge. This paper describes a novel framework for robots to incrementally learn object models and categorize objects based on mul...
متن کاملTowards Robust Human-Robot Interaction using Multimodal Cues
Real-world domains characterized by partial observability and nondeterminism frequently make it difficult for a robot to operate without any human feedback. However, human participants are unlikely to have the time and expertise to provide elaborate and accurate feedback. The deployment of mobile robots to interact with humans in dynamic domains hence requires that the robot learn from multimod...
متن کاملTurn Taking for Human-Robot Interaction
Applications in Human-Robot Interaction (HRI) in the notso-distant future include robots that collaborate with factory workers or serve us as caregivers or waitstaff. When offering customized functionality in these dynamic environments, robots need to engage in real-time exchanges with humans. Robots thus need to be capable of participating in smooth turn-taking interactions. The research goal ...
متن کاملCynthia Breazeal Regulation and Entrainment in Human - Robot Interaction
Newly emerging robotics applications for domestic or entertainment purposes are slowly introducing autonomous robots into society at large. A critical capability of such robots is their ability to interact with humans, and in particular, untrained users. In this paper we explore the hypothesis that people will intuitively interact with robots in a natural social manner provided the robot can pe...
متن کامل