The primary task of marine surveillance is to construct a perfect marine situational awareness (MSA) system that serves to safeguard national maritime rights and interests and to maintain blue homeland security. Progress in maritime wireless communication, developments in artificial intelligence, and automation of marine turbines together imply that intelligent shipping is inevitable in future global shipping. Computer vision-based situational awareness provides visual semantic information to human beings that approximates eyesight, which makes it likely to be widely used in the field of intelligent marine transportation. We describe how we combined the visual perception tasks required for marine surveillance with those required for intelligent ship navigation to form a marine computer vision-based situational awareness complex and investigated the key technologies they have in common. Deep learning was a prerequisite activity. We summarize the progress made in four aspects of current research: full scene parsing of an image, target vessel re-identification, target vessel tracking, and multimodal data fusion with data from visual sensors. The paper gives a summary of research to date to provide background for this work and presents brief analyses of existing problems, outlines some state-of-the-art approaches, reviews available mainstream datasets, and indicates the likely direction of future research and development. As far as we know, this paper is the first review of research into the use of deep learning in situational awareness of the ocean surface. It provides a firm foundation for further investigation by researchers in related fields.

The data quality and real-time analysis of automatic identification system (AIS) are of great significance for water transportation safety and intelligent maritime construction. To improve the AIS data quality and analyze AIS data in real time, a real-time AIS data cleaning and indicator analysis algorithm is proposed. This algorithm performs distributed real-time data cleaning and analysis for massive AIS data based on stream computing technology. It includes data fusion, deduplication, decoding, abnormal data identification, sequencing, prediction, and statistics steps. Abnormal AIS data are repaired by linear regression, multiple trajectory tracking, caching, and other technologies. The AIS status is determined in real-time via multidimensional AIS packet loss analyses, multifactor AIS data statistics, and spatial-temporal data visualization, effectively improving the intelligence level of maritime supervision applications. The proposed algorithm has been running on a production environment, and it monitors AIS data in a certain section of the Yangtze River Basin 24 hours every day without interruption. The operation results show that the proposed algorithm can improve the quality of AIS data, addresses ship trajectory jump issues, and provides timely position updates. The real-time indicator analysis results can provide the data support for ship navigation and maritime supervision.

For path planning algorithms based on visibility graph, constructing a visibility graph is very time-consuming. To reduce the computing time of visibility graph construction, this paper proposes a novel global path planning algorithm, bidirectional SVGA (simultaneous visibility graph construction and path optimization by A⁎). This algorithm does not construct a visibility graph before the path optimization. However it constructs a visibility graph and searches for an optimal path at the same time. At each step, a node with the lowest estimation cost is selected to be expanded. According to the status of this node, different through lines are drawn. If this line is free-collision, it is added to the visibility graph. If not, some vertices of obstacles which are passed through by this line are added to the OPEN list for expansion. In the SVGA process, only a few visible edges which are in relation to the optimal path are drawn and the most visible edges are ignored. For taking advantage of multicore processors, this algorithm performs SVGA in parallel from both directions. By SVGA and parallel performance, this algorithm reduces the computing time and space. Simulation experiment results in different environments show that the proposed algorithm improves the time and space efficiency of path planning.

Simultaneous localization and mapping (SLAM) is a typical computing-intensive task. Based on its own computing power, a mobile robot has difficult meeting the real-time performance and accuracy requirements for the SLAM process at the same time. Benefiting from the rapid growth of the network data transmission rate, cloud computing technology begins to be applied in the robotics. There is the reliability problem caused by solely relying on cloud computing. To compensate for the insufficient airborne capacity, ensure the real-time performance and reliability, and improve the accuracy, a SLAM algorithm based on edge-cloud collaborative computing is proposed. The edge estimates the mobile robot pose and the local map using a square root unscented Kalman filter (SR-UKF). The cloud estimates the mobile robot pose and the global map using a distributed square root unscented particle filter (DSR-UPF). By using sufficient particles in the cloud, DSR-UPF can improve the SLAM accuracy. The cloud returns the particle with the largest posteriori probability to the edge, and the edge performs edge-cloud data fusion based on probability. Both the simulation and the experimental results show that the proposed algorithm can improve the estimation accuracy and reduce the execution time at the same time. By transferring the heavy computation from robots to the cloud, it can enhance the environmental adaptability of mobile robots.