Tracking in Uncertain Environments

Abstract

A tracking filter is an algorithm for estimating the state (such as position and velocity) of an object from measurements of a sensor such as a radar. This study concerns the problem of tracking a target when the origin of the sensor measurements is uncertain. The full Bayesian solution to this type of problem gives rise to Gaussian mixture distributions, which are composed of an ever increasing number of components. To implement such a tracking filter, this growth of components must be controlled by approximating the mixture distribution. Two algorithms have been developed for approximating Gaussian mixture distributions. These techniques attempt to minimize the number of mixture components without modifying the 'structure' of the distribution beyond a specified limit. Also the final approximation is itself a Gaussian mixture. The performance of the algorithms has been assessed by simulation for the problem of tracking a single target in the presence of uniformly distributed false measurements. This assessment indicates the significant range of problem parameters where the new algorithms give a substantial performance improvement over the well known Probabilistic Data Association Filter (which approximates the mixtures by a single Gaussian component). The tracking example is extended in the second part of this study to show how the Bayesian approach may be applied to more complex uncertain tracking problems, including that of fusing data from several independent sources. Great Britain.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1989
Accession Number
ADA215866

Entities

People

  • D. J. Salmond

Organizations

  • Royal Aircraft Establishment

Tags

Communities of Interest

  • C4I
  • Materials and Manufacturing Processes
  • Sensors

DTIC Thesaurus Topics

  • Bayes Theorem
  • Bayesian Networks
  • Computational Science
  • Computer Programs
  • Computers
  • Confidence Limits
  • Data Association
  • Data Fusion
  • Filtration
  • Kalman Filters
  • Machine Learning
  • Mathematical Filters
  • Multitarget Tracking
  • Plastic Explosives
  • Probability
  • Target Tracking
  • Two Dimensional

Readers

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

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms