Decision making implications of multi-sensor data fusion

PDF

Authors
  1. Klein, G.
Corporate Authors
Defence and Civil Inst of Environmental Medicine, Downsview ONT (CAN);Klein Associates Inc, Fairborn OH (US)
Abstract
The use of multi-data sensor fusion to support Operations Room Officers in Halifax Class frigates offers an important potential to improve performance in detecting and identifying tracks. However, there is a risk that the fusion algorithms may also interfere with performance. This report used a Naturalistic Decision Making perspective to identify ways that fusion algorithms might create barriers to the development and application of expertise. One specific concern was that the level of uncertainty might not be reduced, but rather would be shifted from one type of uncertainty to another. Operators using fusion algorithms may have to confront uncertainties about the ationale for system outputs, and may have more difficulty in gaining access to critical data. Another concern was that a system that was efficient during normal operations would become inefficient during non-routine and anomalous events. Recommendations were made for additional research on the relationship between fusion algorithms and expertise. This report takes a deliberately critical position on multi-sensor data fusion (MSDF) because of concerns about its effects on expertise. However, the benefits of MSDF are clear, as is the need to find ways to develop strategies that will make fusion algorithms less disruptive and more valuable in improving performance.
Keywords
MSDF (Multi-Sensor Data Fusion);MSDF (Multi-Source Data Fusion);Fusion algorithms;Naturalistic decision making;Individual and team performance;Decision Support Systems (DSS)
Report Number
DCIEM-CR-2000-108 — Contractor Report
Date of publication
15 Sep 2000
Number of Pages
37
DSTKIM No
CA010308
CANDIS No
514823
Format(s):
Hardcopy;Document Image stored on Optical Disk

Permanent link

Document 1 of 1

Date modified: