PACT: Possibilistic Approach to Correlation and Tracking

Abstract

A procedure is presented in this paper for treating the nongeolocational aspects of multi target correlation and tracking based on the well developed calculi of possibility theory and certain new theoretical considerations. More generally, it is shown that a combination of possibilistic and probabilistic techniques leads to both a feasible and optimal parameter estimation algorithm. The technique incorporates observed data, error distributions or equivalently, matching tables for attribute outcome values and inference rules connecting attribute matching intensities with consequent correlation levels.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 10, 1982
Accession Number
ADA241181

Entities

People

  • I. R. Goodman

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence
  • Bayes Theorem
  • Classification
  • Computers
  • Estimators
  • Fuzzy Sets
  • Governments
  • Heuristic Methods
  • Intensity
  • Multiple Targets
  • Notation
  • Optimal Estimators
  • Random Variables
  • Search Theory
  • Theorems
  • Translations

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Regression Analysis.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference