The underwater acoustic target signal is affected by factors such as the underwater environment and the ship’s working conditions, causing the generalization of the recognition model is essential. This study is devoted to improving the generalization of recognition models, proposing a feature extraction module based on neural network and time-frequency analysis, and validating the feasibility of the model-based transfer learning method. A network-based filter based on one-dimensional convolution is built according to the calculation mode of the finite impulse response filter. An attention-based model is constructed using the convolution network components and full-connection components. The attention-based network utilizes convolution components to perform the Fourier transform and feeds back the optimization gradient of a specific task to the network-based filter. The network-based filter is designed to filter the observed signal for adaptive perception, and the attention-based model is constructed to extract the time-frequency features of the signal. In addition, model-based transfer learning is utilized to further improve the model’s performance. Experiments show that the model can perceive the frequency domain features of underwater acoustic targets, and the proposed method demonstrates competitive performance in various classification tasks on real data, especially those requiring high generalizability.

The geomagnetic field constitutes a massive fingerprint database, and its unique structure provides potential position correction information. In recent years, particle filter technology has received more attention in the context of robot navigation. However, particle degradation and impoverishment have constrained navigation systems’ performance. This paper transforms particle filtering into a particle-matching positioning problem and proposes a geomagnetic localization method based on an adaptive fission particle filter. This method employs particle-filtering technology to construct a geomagnetic matching positioning model. Through adaptive particle fission and sampling, the problem of particle degradation and impoverishment in traditional particle filtering is solved, resulting in improved geomagnetic matching positioning accuracy. Finally, the proposed method was tested in a sea trial, and the results show that the proposed method has a lower positioning error than traditional particle-filtering and intelligent particle-filtering algorithms. Under geomagnetic map conditions, an average positioning accuracy of about 546.44 m is achieved.

In order to obtain the measurement parameters of the sea bottom geomorphology or underwater objects, the first step in side-scan sonar (SSS) image processing is bottom detection. Due to the complexity of the marine environment, the acoustic signals received by SSS are usually polluted by noises, which affect its image quality and make the extraction of image features difficult. To address this problem, this study proposes an automatic detection method for the sea bottom line based on the actual experimental acquisition of SSS images, which is supposed to support the autonomous underwater vehicle (AUV) for intelligent target detection and classification. The proposed method comprises four main steps. First, the raw SSS data is analyzed to obtain a grayscale image, and the blind zone boundary of the image is obtained using the threshold method. Then, the noise characteristics of the image are analyzed and the denoising algorithm is optimized to effectively remove high-frequency noise. Next, spatial-temporal matching calculations are performed on each ping port and starboard data, and the accurate coordinates of first bottom returns are obtained through extreme value detection. Finally, automatic and accurate detection of the bottom line is realized according to the smooth processing of the coordinate sequence of first bottom returns. The experiments have demonstrated the effectiveness of the proposed method. As the method does not require human intervention in adjusting parameters during operation, the proposed method with a certain time window imposed during image acquisition will be suitable for AUV missions when the SSS is determined.

Target classification and recognition have always been complex problems in underwater acoustic signal processing because of noise interference and feature instability. In this paper, a robust feature extraction method based on multi-task learning is proposed, which provides an effective solution. Firstly, an MLP-based network model suitable for underwater acoustic signal processing is proposed to optimize feature extraction. Then, multi-task learning is deployed on the model in hard parameter-sharing so that the model can extract anti-noise interference features and embed prior feature extraction knowledge. In the model training stage, the simultaneous training method enables the model to improve the robustness and representation of classification features with the knowledge of different tasks. Furthermore, the optimized classification features are sent to the classification network to complete target recognition. The proposed method is evaluated by the dataset collected in the real environment. The results show that the proposed method effectively improves recognition accuracy and maintains high performance under different noise levels, which is better than popular methods.

Sparse arrays play an important role in the widely used acoustic and electromagnetic applications such as sonar, radar, and satellite. In this paper, a recently proposed swarm intelligence algorithm, salp swarm algorithm (SSA), is introduced for design of both linear and planar sparse antenna arrays with optimum sidelobe level (SLL). The design of sparse antenna arrays can be regarded as a nonlinear optimization problem, which the main optimization goal is finding optimum excitations and positions for antenna elements to achieve antenna array beam pattern with desired SLL. The SSA is a new kind of swarm intelligence algorithm which inspired by the swarming behavior of salps. Compared to other evolutionary algorithms, the SSA requires few algorithm-specified tuning parameters, which makes it simple and easy to use for engineering. Illustrative examples of both linear and planar arrays are given to verify the validity of the SSA for SLL reduction of sparse arrays.

Vector magnetometers play an essential role in aerial and marine navigation due to their advantages of low cost, portability, and high sensitivity. However, the unstable magnetic field anomaly restricts the performance improvement of the magnetic compass. The existing methods have done a lot of optimization and improvement work, but the test results are less than expected in a dynamic environment. In order to solve this problem, a bionic magnetic compass algorithm based on the theory of “radical pairs” in birds was proposed. This paper finds the stability convergence of the magnetic compass deviation (MCD) at the critical angle. The theoretical analysis and experiment results show that the heading accuracy at the critical angle is trustworthy. Then, the MCD distribution and magnetic heading correction model be established. Finally, the current magnetic heading is corrected according to the angle difference between the critical angle and the current heading. The sea trial shows that compared with the traditional compensation algorithm, reservoir environment is improved by 83.1%, and the effect in real ocean environment improved by 30.12% and 58.49%.