Abstract

This paper presents a novel framework for the problem of target localization based on the range information collected by a single mobile agent. The proposed methodology exploits the algebra of Volterra integral operators to annihilate the influence of initial conditions on the transient phase, thus achieving a deadbeat performance. The robustness properties against additive measurement perturbations are analyzed, and the bias caused by the time discretization is characterized as well. Extensive simulation results and comparisons are provided showing the effectiveness of the proposed technique in coping with both stationary and drifting targets.

Highlights

  • The problem of target localization represents one of the fundamental issues in several engineering fields such as aerospace, military, wireless sensor networks, etc

  • The localization systems can be classified into two categories: the global positioning systems (GPS) and the local positioning system (LPS) [2]

  • Volterra integral operators play a key role in transforming a differential expression into a sequence of algebraic equations

Read more

Summary

INTRODUCTION

The problem of target localization represents one of the fundamental issues in several engineering fields such as aerospace, military, wireless sensor networks, etc. A robust correctionbased state estimator is designed with the consideration of disturbances on both the distance measurement and velocity Another family of approaches is based on regression formulation derived from geometrical relationship (see [15], [16], [17], [18]). Besides the aforementioned linear estimation frameworks, in our recent work [21], Volterra integral operators have been applied to the nonlinear range-based localization problem, yielding the position estimation of a fixed target in a deadbeat manner in the ideal case. The robust stability analysis provides evidence that the estimation error remains bounded in case of a persistently drifting target and measurement noise, making the proposed algorithm a viable choice for the practical implementation.

PROBLEM STATEMENT AND PRELIMINARIES
NON-ASYMPTOTIC VOLTERRA OPERATORS ALGEBRA
Dynamic implementation of Volterra operators
Kernel conditions for non-asymptoticity
A KERNEL-BASED NON-ASYMPTOTIC LOCALIZATION ALGORITHM
Stability analysis
Analysis of the bias introduced by the time-discretization of the algorithm
DRIFTING TARGET
SIMULATION RESULTS AND COMPARISONS
Stationary target: estimation under noise-free condition
Stationary target: estimation in noisy scenario
Drifting target: estimation under noise-free condition
Drifting target: estimation in noisy scenario
VIII. CONCLUDING REMARKS
Methods
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call