Abstract
When an oil spill occurs, a prompt response reduces significantly the impact. The preparedness and contingency plans are essential to identify the most appropriate technologies. Unmanned and autonomous vehicles (UAVs) is emerging as a powerful tool of strategic potential in the observation, oil tracking and damage assessment of an oil spill. The SpilLess project explored the suitability of these devices to be the first-line response to an oil spill. This work analyses the operational requirements related to environmental parameters following a two steps approach: 1) Environmental characterization from long wind and waves time series and modelling; 2) Definition of the optimal periods for operating each UAVs. We have defined the periods in which each of these facilities acts best, confirming that the operational limits of UAVs are not significantly more restrictive than the traditional operations. UAVs should be included in contingency plans as available tools to fight against oil spills.

Abstract
Autonomous Surface Vehicles (ASVs), operating near ship harbors or relatively close to shorelines must be able to steer away from incoming vessels and other possible obstacles, be they dynamic or not. To do this, one must implement some type of multi-target tracking and obstacle avoidance algorithms that lets the vehicle dodge obstacles. This paper presents a radar-based multi-target tracking system developed for obstacle detection in a small unmanned surface vehicle. The system was designed for ROAZ II ASV belonging to INESC TEC/ISEP and implemented in Robot Operating System (ROS) for easier integration with the already existing software.

Abstract
This work addresses the use of hyperspectral imaging systems for remote detection of marine litter concentrations in oceanic environments. The work consisted on mounting an off-the-shelf hyperspectral imaging system (400-2500 nm) in two aerial platforms: manned and unmanned, and performing data acquisition to develop AI methods capable of detecting marine litter concentrations at the water surface. We performed the campaigns at Porto Pim Bay, Fail Island, Azores, resorting to artificial targets built using marine litter samples. During this work, we also developed a Convolutional Neural Network (CNN-3D), using spatial and spectral information to evaluate deep learning methods to detect marine litter in an automated manner. Results show over 84% overall accuracy (OA) in the detection and classification of the different types of marine litter samples present in the artificial targets.

Abstract
Offshore wind turbine application has been widespread in the last years, with an estimation that in 2030 will reach a total capacity of 234GW. Offshore wind farms introduce advantages in terms of environmental impact (noise, impact on birds, disrupted landscapes) and energy production (34% onshore and 43% offshore). Still, they also introduce scientific challenges in developing methodologies that allow wind farm inspection (preventive maintenance) safety for humans. This paper presents a UAV approach for autonomous inspection of inland windturbine and describes the field tests in Penela, Portugal. From the state-of-the-art available wind turbine inspection, in 2015, we carried out the first autonomous inspection with a UAV. The inspection of wind blades offshore is an ongoing project; therefore, the paper also presents the preliminary results with a simulation environment to validate the 3D LiDAR and the inspection procedure with new challenges effects: floating platform, wind gusts, and unknown initial blade position.

Abstract
In recent years, Artificial Intelligence (AI) provided essential tools to enhance the productivity of activities related to civil engineering, particularly in design, construction, and maintenance. In this framework, the present work proposes a novel AI computer vision methodology for automatically identifying the corrosion phenomenon on roofing systems of large-scale industrial buildings. The proposed method can be incorporated into computational packages for easier integration by the industry to enhance the inspection activities' performance. For this purpose, a dedicated image database with more than 8k high-resolution aerial images was developed for supervised training. An Unmanned Aerial Vehicle (UAV) was used to acquire remote georeferenced images safely and efficiently. The corrosion anomalies were manually annotated using a segmentation strategy summing up 18,381 instances. These anomalies were identified through instance segmentation using the Mask based Region-Convolution Neural Network (Mask R-CNN) framework adjusted to the created dataset. Some adjustments were performed to enhance the performance of the classification model, particularly defining an adequate input image size, data augmentation strategy, Intersection over a Union (IoU) threshold during training, and type of backbone network. The inferences show promising results, with correct detections even under complex backgrounds, poor illumination conditions, and instances of significantly reduced dimensions. Furthermore, in scenarios without a roofing system, the model proved reliable, not producing any false positive occurrences. The best model achieved metrics' values equal to 65.1% for the bounding box detection Average Precision (AP) and 59.2% for the mask AP, considering an IoU of 50%. Regarding classification metrics, the precision and recall were equal to 85.8% and 84.0%, respectively. The developed methodology proved to be extremely valuable for guiding infrastructure managers in taking physically informed decisions based on the real assets condition.

Abstract