Steps toward detecting and recovering from perceptual failures
Mathias Broxvall, Lars Karlsson, Alessandro Saffiotti · 2004
An important requirement for autonomous systems is the ability to detect and recover from exceptional situations such as failures in observations. In this paper we investigate how traditional AI planning techniques can be used to reason about observations and to recover from these situations. In this first step we concentrate on failures in perceptual anchoring. We illustrate our approach by showing experiments run on a mobile robot equipped with a color camera.