In recent years, the use of Unmanned Aerial Systems (UAS) has become commonplace in a wide variety of tasks due to their relatively low cost and ease of operation. In this paper, we explore the use of UAS in maritime Search And Rescue (SAR) missions by using experimental data to detect and classify objects at the sea surface. The objects are chosen as common objects present in maritime SAR missions: a boat, a pallet, a human, and a buoy. The data consists of thermal images and Gaussian Mixture Model (GMM) is used to discriminate foreground objects from the background. Then, bounding boxes containing the object are defined and used to train a Convolutional Neural Network (CNN). The CNN achieves the average accuracy of 92.5% when evaluating a testing dataset.

Unmanned Aerial Vehicles (UAVs) have recently been used in a wide variety of applications due to their versatility, reduced cost, rapid deployment, among other advantages. Search and Rescue (SAR) is one of the most prominent areas for the employment of UAVs in place of a manned mission, especially because of its limitations on the costs, human resources, and mental and perception of the human operators. In this work, a real-time path-planning solution using multiple cooperative UAVs for SAR missions is proposed. The technique of Particle Swarm Optimization is used to solve a Model Predictive Control (MPC) problem that aims to perform search in a given area of interest, following the directive of international standards of SAR. A coordinated turn kinematic model for level flight in the presence of wind is included in the MPC. The solution is fully implemented to be embedded in the UAV on-board computer with DUNE, an on-board navigation software. The performance is evaluated using Ardupilot’s Software-In-The-Loop with JSBSim flight dynamics model simulations. Results show that, when employing three UAVs, the group reaches 50% Probability of Success 2.35 times faster than when a single UAV is employed.

This work presents a novel framework providing the ability to control an Unmanned Aerial System (UAS) while detecting objects in real-time with visible detections, containing class names, bounding boxes, and confidence scores, in a changeable high-fidelity sea simulation environment, where the major attributes like the number of human victims and debris floating, ocean waves and shades, weather conditions such as rain, snow, and fog, sun brightness and intensity, camera exposure and brightness can easily be manipulated. Developed using Unreal Engine, Microsoft Air-Sim, and Robot Operating System (ROS), the framework was firstly used to find the best possible configuration of the UAS flight altitude, and camera brightness with high average prediction confidence of human victim detection, and then only autonomous real-time test missions were carried out to calculate the accuracies of two pretrained You Only Look Once Version 7 (YOLOv7) models: YOLOv7 retrained on SeaDronesSee Dataset (YOLOv7-SDS) and YOLOv7 originally trained on Microsoft COCO Dataset (YOLOv7-COCO), which resulted in high values of 97.8% and 93.79%, respectively. Furthermore, it is proposed that the framework developed in this study can be reverse engineered for autonomous real-time training with automatic ground-truth labeling of the images from the gaming engine that already has all the details of all objects placed in the environment for rendering them onto the screen. This is required to be done to avoid the cumbersome and timeconsuming manual labeling of large amount of synthetic data that can be extracted using this framework which could be a groundbreaking achievement in the field of maritime computer vision.