Acoustic Fingerprint Recognition Using Artificial Neural Networks

This study presents the modeling and control of the unmanned surface vehicle (USV) SABALO. Two models were built, one based on a transfer function matrix and another based on state variables, and from these models, two control strategies were developed. The first strategy is based on independent Proportional-Integral/Proportional-Derivative (PI/PD) controllers complemented by a decoupling system, and the second strategy is based on state variable feedback. The two control strategies were evaluated and contrasted. Results demonstrated that the decoupler effectively eliminated variable interaction, enhancing stability in straight trajectories and directional changes. Meanwhile, state feedback control demonstrated markedly faster response times and superior precision, accompanied by higher energy consumption. The study concludes that both strategies are effective, but their suitability depends on the mission. The decoupler could be ideal for energy-efficient, long-duration operations, while state feedback could be appropriate for dynamic environments requiring rapid maneuvers.

This paper presents a technique, based on the matrix pencil method (MPM), for the compression of underwater acoustic signals produced by boat engines. The compressed signal, represented by its complex resonance expansion, is intended to be sent over a low-bit-rate wireless communication channel. We demonstrate that the method can provide data compression greater than 60%, ensuring a correlation greater than 93% between the reconstructed and the original signal, at a sampling frequency of 2.2 kHz. Once the signal was reconstituted, a localization process was carried out with the time reversal method (TR) using information from four different sensors in a simulation environment. This process sought to achieve the identification of the position of the ship using only passive sensors, considering two different sensor arrangements.

This paper presents the results obtained from the design of a speed and heading control system for the platform component of an unmanned surface vehicle (USV) technology demonstrator, using PID controllers. A description of the vehicle technology demonstrator used for the implementation of the system is presented, as well as the tests performed to obtain the dynamic models of the platform component of the USV technology demonstrator, the calculations of the controllers and the navigation tests that validate the performance of these controllers. All the above as a result of the joint research process between the Naval Cadet School "Almirante Padilla" and COTECMAR, with the support of the Naval NCO School "ARC BARRANQUILLA" and financed by the Colombian Navy through Minciencias.

Deep learning in cloud environments offers benefits such as scalability, cost-effectiveness, and flexibility, yet presents several challenges. These include managing large datasets, allocating resources, and synchronizing processes across multiple machines. Training deep learning models on the cloud requires careful consideration of scalability, data storage and transfer, resource allocation, and model management. Overfitting and computation time also pose challenges, often addressed through regularization methods, dropout techniques, and optimization of training parameters. Techniques to optimize deep learning in the cloud include using GPUs and specialized deep learning processors like TPUs to speed up computation. Efficient data storage methods, such as those used with Neural Concept Shape, help manage large 3D numerical simulation files. Optimization strategies focus on improving computational efficiency and reducing costs. Deep learning applications in cloud environments span diverse fields, including image reconstruction, weather prediction, and financial mathematics. Future trends involve exploring hybrid forecasting models, investigating the impact of emerging technologies, and developing innovative methods for handling missing data to enhance predictive accuracy and robustness. The use of integrated photonic hardware accelerators and atomically thin semiconductors also shows promise for energy-efficient deep learning