Abstract
Solar wind forecasting is a core component of Space Weather, a field that has been the target of many novel machine-learning approaches. The continuous monitoring of the Sun has provided an ever-growing ensemble of observations, facilitating the development of forecasting models that predict solar wind properties on Earth and other celestial objects within the solar system. This enables us to prepare for and mitigate the effects of solar wind-related events on Earth and space. The performance of some simulation-based solar wind models depends heavily on the quality of the initial guesses used as initial conditions. This work focuses on improving the accuracy of these initial conditions by employing a Recurrent Neural Network model. The study's findings confirmed that Recurrent Neural Networks can generate better initial guesses for the simulations, resulting in faster and more stable simulations. In our experiments, when we used predicted initial conditions, simulations ran an average of 1.08 times faster, with a statistically significant improvement and reduced amplitude transients. These results suggest that the improved initial conditions enhance the numerical robustness of the model and enable a more moderate integration time step. Despite the modest improvement in simulation convergence time, the Recurrent Neural Networks model's reusability without retraining remains valuable. With simulations lasting up to 12 h, an 8% gain equals one hour saved per simulation. Moreover, the generated profiles closely match the simulator's, making them suitable for applications with less demanding physical accuracy.

Abstract
Composites of poly(vinyl alcohol) (PVA) in the shape of braids, in combination with crystals of hydroxyapatite (HAp), were analyzed to perceive the influence of this bioceramic on both the quasi-static and viscoelastic behavior under tensile loading. Analyses involving energy-dispersive X-ray spectroscopy (EDS) and scanning electron microscopy (SEM) allowed us to conclude that the production of a homogeneous layer of HAp on the braiding surface and the calcium/phosphate atomic ratio were comparable to those of natural bone. The maximum degradation temperature established by thermogravimetric analysis (TGA) showed a modest decrease with the addition of HAp. By adding HAp to PVA braids, an increase in the glass transition temperature (Tg) is noticed, as demonstrated by dynamic mechanical analysis (DMA) and differential thermal analysis (DTA). The PVA/HAp composite braids' peaks were validated by Fourier transform infrared (FTIR) spectroscopy to be in good agreement with common PVA and HAp patterns. PVA/HAp braids, a solution often used in the textile industry, showed superior overall mechanical characteristics in monotonic tensile tests. Creep and relaxation testing showed that adding HAp to the eight and six-braided yarn architectures was beneficial. By exhibiting good mechanical performance and most likely increased biological qualities that accompany conventional care for bone applications in the fracture healing field, particularly multifragmentary ones, these arrangements can be applied as a fibrous fixation system.

Abstract
This article addresses the 3-D localization of a stand-alone acoustic beacon based on the Principle of Synthetic Baseline using a single receiver on board a surface vehicle. The process only uses the passive reception of an acoustic signal with no explicit synchronization, interaction, or communication with the acoustic beacon. The localization process exploits the transmission of periodic signals without synchronization to a known time reference to estimate the time-of-arrival (ToA) with respect to an absolute time basis provided by the global navigation satellite system (GNSS). We present the development of the acoustic signal acquisition system, the signal processing algorithms, the data processing of times-of-arrival, and an estimator that uses times-of-arrival and the coordinates where they have been collected to obtain the 3-D position of the acoustic beacon. The proposed approach was validated in a real field application on a search for an underwater glider lost in September 2021 near the Portuguese coast.

Abstract
Accurate localization is a critical task in underwater navigation. Typical localization methods use a set of acoustic sensors and beacons to estimate relative position, whose geometric configuration has a significant impact on the localization accuracy. Although there is much effort in the literature to define optimal 2D or 3D sensor placement, the optimal sensor placement in irregular and constrained 3D surfaces, such as autonomous underwater vehicles (AUVs) or other structures, is not exploited for improving localization. Additionally, most applications using AUVs employ commercial acoustic modems or compact arrays, therefore the optimization of the placement of spatially independent sensors is not a considered issue. This article tackles acoustic sensor placement optimization in irregular and constrained 3D surfaces, for inverted ultra-short baseline (USBL) approaches, to improve localization accuracy. The implemented multi-objective memetic algorithm combines an evaluation of the geometric sensor's configuration, using the Cramer-Rao Lower Bound (CRLB), with the incidence angle of the received signal. A case study is presented over a simulated homing and docking scenario to demonstrate the proposed optimization algorithm.

Abstract
There are several compelling reasons for exploring the ocean, for instance, the potential for accessing valuable resources, such as energy and minerals; establishing sovereignty; and addressing environmental issues. As a result, the scientific community has increasingly focused on the use of autonomous underwater vehicles (AUVs) for ocean exploration. Recent research has demonstrated that buoyancy change modules can greatly enhance the energy efficiency of these vehicles. However, the literature is scarce regarding the dynamic models of the vertical motion of buoyancy change modules. It is therefore difficult to develop adequate depth controllers, as this is a very complex task to perform in situ. The focus of this paper is to develop simplified linear models for a buoyancy change module that was previously designed by the authors. These models are experimentally identified and used to fine-tune depth controllers. Experimental results demonstrate that the controllers perform well, achieving a virtual zero steady-state error with satisfactory dynamic characteristics.

Abstract
This paper describes a robotic system to detect and estimate the volume of sediments in underwater wall corners, in scenarios with zero visibility. All detection and positioning is based on data from a scanning sonar. The main idea is to scan the walls and the bottom of the structure to detect the corner, and then use data obtained in the direction of the corner to estimate the presence of sediment accumulation and its volume. Our approach implements an image segmentation to extract range from the surfaces of interest. The resulting data is then employed for relative localization and estimate of the sediment accumulation. The paper provides information about the methodologies developed and data from practical experiments.