Abstract
By "intelligently" locating a sensor with respect to its envi ronment, it is possible to minimize the number of sensing operations required to perform many tasks. This is particu larly important for sensing media, such as tactile sensors and sonar, that provide only "sparse" data. In this paper, a sys tem is described that uses the principles of statistical decision theory to determine the optimal sensing locations for per forming recognition and localization operations. The system uses a Bayesian approach to utilize any prior object informa tion (including object models or previously acquired sensory data) in choosing the sensing locations.
Get full access to this article
View all access options for this article.
