Traditional maritime surveillance often relies on static cameras, limiting coverage and efficiency in detecting vessels. This paper proposes a novel real-time UAV-based vessel detection system that addresses these limitations. The system leverages a Dynamic Camera Control Strategy that overcomes fixed field-of-view constraints by adjusting camera angles based on predefined search patterns, historical data, and real-time sensor feedback. This systematic scanning approach ensures comprehensive coverage, minimizing missed detections. Furthermore, a Feature-Based Prioritization Scheme facilitates real-time target confirmation by analyzing features like size, shape, and, if applicable from additional sensors, other relevant data. This scheme prioritizes promising candidates for further analysis, reducing false positives. By comparing features with a reference vessel stored in the system (using a ResNet50-based module), the system effectively discriminates between vessels and other objects, while movement analysis helps distinguish stationary objects. Our system was evaluated using real-world testing and simulation, demonstrating significant improvements in detection accuracy and processing speed compared to state-of-the-art methods. This research presents a valuable contribution to UAV-based maritime surveillance, enhancing operational efficiency and detection accuracy for various real-world applications.

Marine intervention tasks for retrieving large and heavy objects from vessels at sea is significantly important for maritime security. However, it remains a challenging task due to the lack of effective retrieval methods. Harsh open-sea conditions, unreliable GNSS due to either interruptions or deliberate spoofing by the vessels and the limitations of single robots in localizing and manipulating these objects pose significant challenges for the operations. To address these challenges, this article proposes an autonomous heterogeneous multi-robot system comprising two UAVs and a 6-DOF robot arm mounted on a USV to perform this complex task in a GNSS-denied marine environment. GNSS limitations for UAV localization are addressed using UWB beacons specially installed on the USV. Our approach involves the robot arm initially estimating the object’s position. Then the UAVs fly over the target vessel and search for the object using this information. Once object is located, a custom-designed tethered bar mechanism attached to the UAVs drags the object towards the workspace of the robot arm mounted in the USV. This work also introduces a decentralized framework to facilitate effective information sharing and collaboration among the robots. Field trials in maritime conditions demonstrate that by combining these elements, our system possesses significant potential for real-world maritime security and rescue applications. Videos for the field trails can be found at https://youtu.be/a_JzZf7QqWM

Detecting maturity of fruits and vegetables, especially avocados, is a critical task in modern agriculture and supply chain management. Moreover, the accurate assessment of maturity can improve the harvesting time and ensure consistent quality for consumers through the supply chain process. A key approach to achieving this is the non-destructive estimation of produce quality. Vision-Based Tactile Sensing (VBTS) technologies, which mimic human tactile perception, offer a novel approach to address this challenge. This paper focuses on the use of two notable VBTS technologies, GelSight and Facebook's DIGIT sensor. Using these technologies, we developed two novel datasets that assess the avocado maturity using the transformer models, marking a novel contribution in this area. We adapted several transformer architectures to the task, conducting experiments on both image classification and regression to estimate avocado firmness. Among the variants tested, the PoolFormer displayed notable results with accuracy of 92% in detecting avocado maturity level when used with tactile data. The datasets and code used in this study will be shared at this URL. Index terms-Vision-based tactile sensors (VBTS), Vision Transformer (ViT), self-attention block, maturity classification.