Autonomous unmanned underwater vehicles (UUVs) play a vital role in diverse underwater operations; localization is of great interest for UUVs mirroring the trend seen in self-driving surface and aerial vehicles. Unlike their land and aerial counterparts, underwater environments lack reliable Global Navigation Satellite Systems (GNSS) due to radio wave attenuation in water. Hence, alternative localization methods are imperative for both navigation and operational purposes. This study thoroughly reviews sensor technologies for underwater localization, including sonar, Doppler velocity log, cameras, and more. Different operations necessitate distinct localization accuracies and vehicle and sensor choices. Environmental factors, such as turbidity, waves, and sound disturbances, impact sensor performance. Conclusions are given on the coincidence between operational requirements and sensor specifications, with special attention to the open concerns. These considerations include aspects such as the line of sight for acoustic positioning systems and the requirement for a feature-rich environment for visual sensors. Lastly, a prediction for the future of underwater localization is given, where the tendencies indicate lower costs for sensors, making operation-specific vehicles more attractive, which aligns with an increased demand for cost-efficient autonomous offshore operations.
Due to increased numbers of offshore structures and subsea cables, there is a high demand for underwater maintenance and monitoring. Common options to meet this demand are sonar mapping and imaging. Sonar mapping provides a reliable way for object detection with a high penetration depth, but it is not suitable for tasks that require a detailed insight into the material composition and color of the object. Imaging can provide in-depth, comprehensive information on material properties and external features. This makes it reasonable to investigate its use for object segmentation. Hyperspectral imaging is a subset of imaging which proved to be more effective for airborne object segmentation compared to RGB imaging. This stems from the fact that hyperspectral imaging contains a higher number of spectral bands, justifying the investigation of its applicability in underwater environments. However, underwater imaging faces major challenges such as a variable data quality which is strongly affected by water turbidity, color distortion and a narrow wavelength transmission window. Most of the prior studies conducted on underwater object segmentation relied on RGB images, such as the work carried out by AAU Energy on object segmentation relying on synthetic data [1]. The applicability of hyperspectral reliant object segmentation underwater is yet to be conclusively defined, however, the promising results obtained in airborne conditions are an encouraging prospect. The contribution of this paper is to investigate the applicability of hyperspectral data for underwater object segmentation. In particular, a segmentation algorithm, evaluated in an artificial environment, was researched.
Underwater radiated noise (URN) from ship propellers has attracted increasing interest in recent years due to its adverse environmental effects on marine life and their communication channels. The environmental concern to reduce shipping noise and the industrial requirements for faster computational tools are driving factors that promote research in the specialized domain of hydroacoustics. This thesis deals with the development of such a computationally efficient numerical tool, which can be used in the prediction of underwater radiated noise in the early design phase of propellers.
The numerical model is developed with two major objectives – versatility in assessing the relative contributions from the major propeller-noise generating mechanisms, and rapidity in prediction of overall noise behaviour. It uses the Farassat-1A solid-FWH formulation of the Ffowcs-Williams- Hawkings equation by defining equivalent acoustic sources on the propeller blade, sheet cavity and tip vortex cavity surfaces. In particular, the application of the solid-FWH formulation to the tip vortex cavity model is the major novelty in this thesis.
The hydrodynamic flow solution is obtained from a potential flow based solver ESPPRO, which includes analytical models of sheet cavitation and tip vortex cavitation. The hydroacoustic numerical model developed within this thesis, DoLPHiN, is a Python-based code that is primarily designed to accept input from ESPPRO; but during the research, the code has also been adapted to read input from the commercial, finite-volume-based Navier-Stokes solver, STAR-CCM+.
The numerical model implementations are verified through analytical case studies for simple geometrical shapes, such as a pulsating sphere and an oscillating cylindrical cavity. The verification study is further extended for propeller geometries by identifying approximate reference solutions in simplified operating conditions. The numerical tool is validated for industrial application through comparison of its noise prediction with model-scale and full-scale noise measurements. Specific characteristics of the propeller noise spectrum are identified in order to evaluate its noise prediction capabilities. The uncertainty factors involved when validating with experimental measurements are also explored in detail. Furthermore, a design study is presented, which shows potential use of the numerical tool in practical propeller design and optimization applications.
Determination of coverage and thickness of marine growth is a useful tool for determining structural loads and drags on marine structures and ships. In this work, we present an algorithmic program based on sonar and optical camera measurements, that estimates both the coverage and thickness of marine-fouling on off-shore structures. The marine-fouling composition is estimated using a Deep-Neural Network, trained using supervised methods, which can distinguish between hard/soft fouling species and the background water and structural components. The marine-fouling thickness is estimated using an HF Forward Looking Sonar, which is applied as a sensitive ultrasonic thickness gauge, when combined with a thickness measurement algorithm. Combined the measurements provide a localized estimate of the marine-fouling coverage and loadings across the structural surfaces, which can be used for automatic inspection evaluation and mission planning.