Experiments in the control of perception in multi-sensor system

Abstract

The paper presents a new approach to management of multiple sensors and perception algorithms in a multi-sensor robotic system. The approach involves real-time selection of process monitors by a Sensory Perception Controller. The selection is based on the minimization of the expected cost of perception with constraints on the uncertainty of perception. The effectiveness and usefulness of the approach is evaluated through experiments involving a range of sensing modalities which may typically be encountered in robotic applications.