Task and environment-sensitive tracking

Abstract

In a mobile robot, visual tracking, like other visual behaviors, takes placeinacontext that includes aspects of the task, the object being tracked, and the background. In this work, prior knowledge of those task and target characteristics that either enable or hinder di#erent real-time image-tracking algorithms, together with run-time evaluation of the robot's environment, are used to select an algorithm appropriate to the context. 1 Introduction The utility of di#erent cues for distinguishing an object from its background, and therefore as the basis for tracking, can vary tremendously. The following examples typify the cues and contexts that we seek to exploit: # Motion cues are more e#ective from a stationary platform than one that is translating, or worse yet, rotating. # Color cues are insensitive to platform motion. # The reliability of color cues can be predetermined by comparing the model to the background. # Motion cues are useful for animate objects, who are rarely co...