Email Document Reference

Enter your email address below and the reference for this document will be sent to shortly from webmaster@ceesi.com.

Title: When Is More Not Better?
Author: Karen T. Sutherland
Source: 1997 Measurement Science Conference
Year Published: 1997
Abstract: It is often assumed that when sensing in autonomous robot navigation, the more measurements taken, the better. In many cases, this is indeed true. However, we have found that in an outdoor, unstructured environment where the number of possible measurements one can take is limited and error is not normally distributed, additional measurements may cause more harm than good. These rogue measurements are not interpreted as outliers due to the fact that there is no observed central tendency in the measurements as a whole. This paper describes ongoing work in determining how a qualitative rating can be used with sparse data to weight measurements and how these weighted measurements can best be combined with map data to produce a good estimate of location. Experiments have been run using U. S. Geographical Survey Digital Elevation Map (DEM) data from mountainous terrain in eastern Utah.

In order to prevent spam and automated file downloads for documents within the Measurement Library, please follow the instructions below and then you will be able to email a reference to this article.