Underwater acoustic source localization is formulated as a feature learning problem within a machine learning framework, where a data-driven approach directly extracts source distance features from hydroacoustic signals. This study systematically compares the localization performance of four machine learning models—decision tree (DT), random forest (RF), support vector machine (SVM), and feedforward neural network (FNN) models—in both classification and regression tasks. Experimental results demonstrate that, in classification tasks, all algorithms achieve effective localization under high signal-to-noise ratio (SNR) conditions, while the DT model exhibits significant noise sensitivity in low-SNR scenarios; regression tasks show reduced model convergence overall, with only the SVM and RF models maintaining basic localization capabilities at a high SNR. For two-dimensional localization, machine learning classification algorithms are employed, revealing systematic accuracy degradation compared to one-dimensional scenarios, where only the RF and SVM models demonstrate practical value under high-SNR conditions. Validation using measured data from the SWellEx-96 experiment’s S5 event confirms that when constructing datasets with frequency-domain acoustic pressure features from the final 35 min segment, the classification task-driven DT, RF, and SVM models all demonstrate reliable localization performance, benefiting from the inherent high-SNR characteristics of the data.

Effective denoising can ensure fast and accurate target detection. This paper presents an electric field measurement system based on a high-speed motion platform, which was built to analyze the characteristics of low frequency electric field noise. An offshore test has shown that it is possible to detect a low-frequency electric field using a high-speed motion platform. Low frequency electric field noise was then collected to analyze its characteristics in terms of time and frequency domains. Based on the noise characteristics, complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) was improved and combined with an adaptive threshold algorithm for denoising and reconstructing target containing noise signals. As revealed in the results, the proposed algorithm achieved highly effective denoising to overcome the line spectrum detection failure resulting from a high-speed motion platform. The detection range had also been improved from the original 853 m to 1306 m, a 53.1% increase.

To analyze the characteristics of the aeromagnetic scalar gradient detection method, a uniformly magnetized ellipsoid is used to simulate an unexploded ordnance, and a magnetic field detection model is established in the International Geomagnetic Reference Field based on rotation matrices. Furthermore, the spatial distribution of the target’s magnetic field is simulated. The results indicate that the scalar gradient detection curve is closely related to the unmanned aerial vehicle (UAV) heading, geomagnetic direction, and target attitude. According to the measured data, the aeromagnetic detection system exhibits differences in the detection of different headings, indicating that some “blind areas” exist in the scalar gradient magnetic detection method. The experimental measurement by a quadrotor UAV equipped with two optical pump magnetometers verifies that the scalar gradient detection method can effectively eliminate the geomagnetic field as well as the interferences of the UAV itself. Furthermore, the angular relationship between the target magnetic field contour distribution and the heading is found to be the main reason that the scalar gradient detection system enters the “blind detection area.” Therefore, a flight strategy of “positive direction + orthogonal grid” is proposed. This method effectively reduces the missed detection rate of scalar gradient detection and provides strategic guidance for the detection path of aeromagnetic scalar gradient system.

Underwater gliders are a new type of unmanned underwater vehicle, characterized by high energy efficiency, long endurance, and low operational costs. They hold broad application prospects in fields such as ocean exploration, resource surveying, maritime surveillance, and military defense. This paper takes underwater gliders as the research subject, analyzing the characteristics of magnetic interference signals under different operational conditions. The study found that during full operational states, the motor’s operation generates interference signals at 17 Hz; during attitude adjustment, the movement of the moving block generates significant interference magnetic fields, especially during the forward and backward motion of the block, where interference signals at 20 Hz are particularly pronounced. To meet the objective of equipping underwater gliders with magnetic field sensors for underwater target detection, this paper proposes an adaptive filtering method based on the Recursive Least Squares (RLS) algorithm. The experimental results indicate that after filtering with the RLS algorithm, the amplitude of the noise signal has been reduced by over 60%, and it can effectively eliminate the noise components at 17 Hz and 20 Hz caused by the glider’s motor. This algorithm achieves an average increase in the signal-to-noise ratio (SNR) of 12 dB, which is equivalent to an approximately 80% improvement in accuracy. It significantly enhances the stability and signal-to-noise ratio of the magnetic field signals of underwater targets. This provides a feasible solution for equipping underwater gliders with magnetic field sensors for underwater target detection, holding important practical engineering significance.

A mathematical model for an electric field induced by an angularly oscillating magnetic dipole was proposed with magnetic vector potential to analyze the characteristics of the electric field induced by a rotating and angularly oscillating magnetic object. This mathematical model was constructed for the electric field induced by a magnetic object oscillating at a certain angle. On this basis, the phase relationship among the three components of the induced electric field was analyzed (defining the right-hand Cartesian coordinate system). Evidently, a phase difference of π/2 always existed between the horizontal components of the electric field induced by a magnetic dipole rotating around the z-axis. The phase difference between the vertical and transverse components in the x–z plane was also π/2. A phase difference of π was observed in the y–z plane. The above theoretical analysis was verified through simulation and experiment. The results showed that the frequency of the induced electric field was related to the angular velocity and angle of rotation. The amplitude was associated with the magnetic moment and the angular velocity and angle of oscillation. The maximum amplitude did not exceed the amplitude of the electric field induced by a magnetic object angularly oscillating at the same velocity. With regard to the amplitude and phase relationship, the three components of the induced electric field measured in the experiment were consistent with the results of the theoretical analysis.

A wave glider is an ocean observation platform that utilizes wave energy to drive and solar energy to power. Its metal structure will generate related electromagnetic fields due to corrosion and underwater motion. In the detection of weak electromagnetic field signals underwater, its own electromagnetic field characteristics will have an impact on signal detection. To study the applicability of electric field sensors and magnetic field sensors on wave glider platforms, the structural characteristics of the wave glider were analyzed, and the installation positions of electric field sensors and magnetic field sensors were designed based on the different motion states of the water surface mother body and underwater towing body. The measured electromagnetic field data of the wave glider platform were measured, and the measured data were analyzed. It was determined that the interference electric field energy under typical working conditions of the wave glider was mainly concentrated within 1 Hz, which decreased with increasing frequency, and the magnitude was mV/m. The magnitude of the interference magnetic field is several tens of nT, indicating that the electromagnetic field interference is significant during the working state of the wave glider. Installing an electric field sensor directly at the bottom of the wave glider will cause significant noise interference, while installing the magnetic field sensor directly at the bottom of the tractor will affect the servo and the shaking-induced magnetic field. Moreover, wave gliders should not use electric field signals below 1 Hz as signal sources, but they can utilize axial frequency electromagnetic fields to detect weak electromagnetic signals underwater.