Active Electro-Location of Objects in the Underwater Environment Based on the Mixed Polarization Multiple Signal Classification Algorithm

Yidong Xu, Wenjing Shang, Lili Guo, Junwei Qi, Yingsong Li, Wei XueSensors 2018 · DOI

TL;DR

This paper introduces a novel active electro-location system for underwater target detection using a Uniform Circular Antenna (UCA) and a Mixed Polarization Multiple Signal Classification (MP-MUSIC) algorithm. By leveraging the Boundary Element Method (BEM) to model target induction fields and a specially designed UCA that only senses induced fields, the system achieves accurate 3D localization without needing prior target size information.

What Problem Does This Paper Address?

Underwater localization is a critical challenge with applications ranging from robotics to deep-sea exploration. Traditional acoustic methods are hampered by environmental factors like temperature variations and noise, while light-based methods fail in dark or turbid conditions. Inspired by the electric sense of fish, this paper aims to develop a robust and accurate electro-location system for underwater environments, capable of detecting both metallic and insulating targets. The core problem is to precisely pinpoint the 3D location of these targets using electrical signals in a challenging underwater medium.

Methods

The proposed method combines two key techniques:

  • Boundary Element Method (BEM): BEM is used to accurately model the electric field induced by a target (metallic or insulating) when subjected to an external electric field. This method discretizes the target's boundary and solves a matrix equation to determine the field distribution.
  • Mixed Polarization Multiple Signal Classification (MP-MUSIC): This is a high-resolution subspace-based algorithm adapted for underwater electro-location. Unlike traditional MUSIC algorithms, MP-MUSIC can handle signal polarization, making it suitable for this application. It estimates the target's 3D position by analyzing the voltage data from the UCA's receiving electrodes.

A key innovation is the Uniform Circular Antenna (UCA) design. It features a central electric dipole source and surrounding receiving electrodes positioned at equipotential points relative to the source. This configuration ensures that the receiving electrodes primarily capture the induction field generated by the target, minimizing interference from the direct field of the source. This simplifies the data processing as it doesn't require measuring field components in different directions.

The paper also explores a simplified model that approximates the target as an equivalent dipole, reducing computational burden and eliminating the need for prior target size information, though at the cost of some accuracy.

Datasets

Not discussed in the paper. The paper relies on simulated data generated using CST Studio Suite and experimental data collected from a custom-built UCA system.

Key Insights and Ideas

  • Induction Field Focus: Designing the UCA to specifically capture the target's induced field, rather than the direct source field, is crucial for simplifying the localization problem and improving accuracy.
  • MP-MUSIC for Underwater: Adapting the MP-MUSIC algorithm, which accounts for signal polarization, proves effective for 3D underwater electro-location, outperforming canonical MUSIC in certain scenarios.
  • BEM vs. Simplified Model Trade-off: The BEM model offers higher accuracy but requires prior knowledge of the target's size. The simplified model is more flexible (no prior knowledge needed) and computationally efficient, making it suitable for real-time applications, but with a slight reduction in accuracy.
  • Electrode Configuration Matters: The optimal number of receiving electrodes for the UCA system was found to be four, with performance degrading beyond this for both BEM and simplified models.

Key Results

  • Accurate Localization: Both simulation and experimental results demonstrate accurate 3D localization of metallic and insulator targets. For example, in simulations, estimated location errors were as low as 0.003 m.
  • Performance Comparison: The BEM model generally shows better performance than the simplified and Rasnow models, especially at lower SNRs or when the target is further from the center. However, the simplified model and Rasnow model perform comparably and better than BEM in specific ranges (0.3 ≤ x ≤ 0.7).
  • Robustness to Noise: The BEM model shows significant improvement in RMS error as SNR increases, indicating robustness.
  • Azimuth Insensitivity: The BEM model with MP-MUSIC is shown to be relatively insensitive to target azimuth changes.
  • Experimental Validation: Physical experiments confirmed the effectiveness of the proposed method, with maximum location errors of 0.055 m for conductor targets and 0.069 m for insulator targets.

Limitations and Future Work

  • Target Size Prior Information: The BEM model requires prior knowledge of the target's size, which can be a limitation in real-world scenarios where this information is unknown. The simplified model addresses this but sacrifices some accuracy.
  • Performance Near Center: Both BEM and simplified models exhibit unsatisfactory locating performance when the target is very close to the center of the UCA system.
  • Ideal vs. Real Electrodes: Simulation results might be influenced by the fact that CST's electrodes are not perfectly ideal, and mesh errors can introduce field calculation inaccuracies.
  • Future Work: The authors plan to design a new UCA system and further improve the locating method to achieve even more accurate results.

Summary updated

Abstract

This article proposes a novel active localization method based on the mixed polarization multiple signal classification (MP-MUSIC) algorithm for positioning a metal target or an insulator target in the underwater environment by using a uniform circular antenna (UCA). The boundary element method (BEM) is introduced to analyze the boundary of the target by use of a matrix equation. In this method, an electric dipole source as a part of the locating system is set perpendicularly to the plane of the UCA. As a result, the UCA can only receive the induction field of the target. The potential of each electrode of the UCA is used as spatial-temporal localization data, and it does not need to obtain the field component in each direction compared with the conventional fields-based localization method, which can be easily implemented in practical engineering applications. A simulation model and a physical experiment are constructed. The simulation and the experiment results provide accurate positioning performance, with the help of verifying the effectiveness of the proposed localization method in underwater target locating.
BibTeX
@inproceedings{xu2018active,
  author = {Yidong Xu and Wenjing Shang and Lili Guo and Junwei Qi and Yingsong Li and Wei Xue},
  title = {{Active Electro-Location of Objects in the Underwater Environment Based on the Mixed Polarization Multiple Signal Classification Algorithm}},
  booktitle = {Sensors},
  year = {2018},
  doi = {10.3390/s18020554},
}
Figure