Optimal Maneuver Detection and Estimation in Multiobject Tracking,

Abstract

This paper presents an algorithm on the multiobject tracking problem within the framework of a hybrid state estimation problem. This permits the construction of the optimal solution to the multiobject tracking problem. However, due to the exponentially growing storage and computational needs of the algorithm with time, some form of suboptimal approximations have to be made. Since these approximations are made within the framework of the optimal algorithm, the nature and trade-offs associated with the approximations can be examined. Furthermore, optimal and suboptimal algorithms for maneuver detection and estimation can also be incorporated within this framework. The paper is organized as follows. Section 2 formulates the problem of multiobject tracking of maneuvering targets within the framework of the hybrid state estimation problem. The computer implementation of the optimal solution is discussed in Section 3. Suboptimal features that reduce the computational requirements are discussed in Section 4. Section 5 we provides some simulation results.

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 1983
Accession Number
ADP002899

Entities

People

  • A. Blitz
  • A. Willsky
  • R. Washburn
  • T. Kurien

Tags

DTIC Thesaurus Topics

  • Algorithms
  • Computers
  • Construction
  • Control Simulators
  • Cooperation
  • Detection
  • Maneuvers
  • Massachusetts
  • Military Research
  • Simulations
  • Simulators
  • Workshops

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.