Data Fusion: Temporal Reasoning and Truth Maintenance
Abstract
This report contains a survey of two techniques that can be used in the field of data fusion: temporal reasoning and truth maintenance. The automatic fusion of intelligence reports necessitates taking into account the factor time. Incoming messages can lead to new interpretations of the current battlefield situation, changing previously made hypotheses. A data fusion system must also be able to make a prediction of what sightings are to be expected, e. g. in the case of columns of vehicles moving past different sensors. This report describes a temporal database system that can capture (some part) of the volatility of the intelligence processing domain. While processing intelligence reports there is always an amount of uncertainty and incompleteness that has to be dealt with. So there is a need for maintaining different lines of reasoning or hypotheses pertaining to the battlefield situation concurrently, and incorporating new information as it becomes available. In this report an assumption-based truth maintenance system provides a framework in which this problem can be solved. A prototype has been developed to demonstrate the applicability of the aforementioned techniques. This prototype, called Mefisto (Modular Environment for Fusion and Interpretation of Sensor data in Tracking Opposing forces), is a simple knowledge-based system integrated with in temporal truth maintenance facility.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1991
- Accession Number
- ADA245600
Entities
People
- A. P. Kenne
- M. Perre
Organizations
- Netherlands Organisation for Applied Scientific Research