Robust sensor fusion for teleoperations

Abstract

The sensor fusion effects architecture for telesystems (teleSFX) is presented. The goal of teleSFX is to reduce the amount of human supervisory control involved in execution, freeing the operator to handle multiple remote robots without sacrificing robustness. The concepts of state-based sensor fusion and intelligent assistance form the foundations for robust sensor fusion at the remote. State-based fusion provides autonomous execution and monitors for sensing anomalies or failures. Intelligent assistance is used to minimize the operator's involvement in the exception-handling process through generation and testing of hypotheses about the cause of the fusion failure and an interactive problem solving interface. Three experiments using data collected from five sensors mounted on a mobile robot are conducted. They demonstrate the detection of sensing failures and the exception-handling activities for a remote robot.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>