We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Minimum error entropy based multiple model estimation for multisensor hybrid uncertain target tracking systems.
- Authors
Li, Shuhui; Feng, Xiaoxue; Deng, Zhihong; Pan, Feng; Ge, Shengyang
- Abstract
In the multisensor target tracking system, the key of the target tracking performance depends on the state estimation accuracy to a great extent. However, the system uncertainties will seriously affect the performance of the state estimation. Up to now, little research focuses on the state estimation for the multi‐sensor hybrid target tracking systems with multiple uncertainties including the multiple models, the unknown inputs and the systematic biases. In this study, the minimum error entropy based on the multiple model estimation for the multisensor hybrid uncertain target tracking systems with the multiple system uncertainties is presented. The minimum variance unbiased filter based on the general systematic bias evolution model decoupled with the unknown state is designed to estimate the optimal systematic biases and compensate the system measurements. Taking full advantage of the compensated measurement information in time and space, the multiple model observer based on the minimum error entropy is designed to obtain the optimal state estimation. The simulation results of the target tracking scenario illustrate the effectiveness of the proposed method, and the indoor target tracking and positioning experiment based on the ultrawideband further verifies that the proposed method is satisfying.
- Publication
IET Signal Processing (Wiley-Blackwell), 2020, Vol 14, Issue 4, p199
- ISSN
1751-9675
- Publication type
Article
- DOI
10.1049/iet-spr.2019.0178