A Vision-based Architecture for Long-Term Human-Robot Interaction

C. King, X. Palathingal, Monica Nicolescu, and Mircea Nicolescu (USA)

Keywords

Tracking, posture recognition, humanrobot interaction.

Abstract

Advances in robotics research bring robots closer to real world applications. Although robots have become increas ingly capable, productive interaction is still restricted to specialists in the field. In this paper, we propose an inter active architecture, based on visual capabilities, which al lows robots to interact with people in a natural way, to deal with multiple users, and to be constantly aware of their sur roundings. First, we endow our robot with visual capabili ties, which allow it to detect when people are requesting to engage it in interaction. Second, we provide the robot with flexibility in dealing with multiple users, such as to accom modate multiple user-requests and task interruptions, over extended periods. The visual capabilities we propose allow the robot to identify multiple users, with multiple postures, in real-time and in dynamic environments, where both the robot and human users are moving. We demonstrate our approach on a Pioneer 3DX mobile robot, performing ser vice tasks in a real-world environment.

Important Links:



Go Back