Unmanned vessels have been ever-increasing in modern fisheries in recent years. Fishery monitoring and supervision can be expected to improve the quality and efficiency with cost savings in smart fisheries, due mainly to their outstanding autonomous navigation and path planning. The promising potential application can also be found in future fishery. In this study, a systematic analysis was provided for the unmanned vessels in intelligent fish farming. Several technologies were given in aquaculture systems, including autonomous navigation, path planning, water environment monitoring, and their integration. Among them, the current status was critically examined to identify the technological gaps for the potential research directions, in order to enhance the efficacy of unmanned vessels in smart fishery. The results show that unmanned vessels significantly enhanced the precision and efficiency of fishery supervision when equipped with high-resolution sensors and intelligent systems. It was very crucial to maintain the real-time monitoring of fishing activities and the automatic identification of illegal fishing practices in the sustainable fishery. Moreover, unmanned vessels were deployed to reduce manual operation, leading to low labor intensity and high overall efficiency. Furthermore, unmanned vessels also shared superior performance in water environment monitoring. The real-time acquisition of environmental data enhanced the monitoring accuracy in aquatic ecosystems. Capital expenditure was also reduced to minimize the extensive investment of equipment under harsh environmental conditions in modern fisheries. The intelligent feeding and medicating technologies were integrated into unmanned vessels, leading to the high operational efficiency in aquaculture. The optimal feeding schedules and precise medication dosages greatly contribute to the better health of aquatic species. Consequently, the overall cost of aquaculture operations was reduced for more sustainable and economically viable farming. Numerous opportunities and challenges were concluded for the broad application of unmanned vessels in fisheries. Several key areas were identified with promising potential applications, including low-cost production, multifunctional integration, new technologies, and innovative scenarios. Moreover, the stability and reliability of unmanned vessels were highlighted in complex and dynamic environmental conditions. It was also required for the integrated multi-parameter detection and the reliable self-powered modules to support long-term autonomous operations. The application scenarios were then extended to deploy the unmanned vessels in a wider scope of smart fisheries.

As UAV technology has become more popular in various fields, tracking UAV swarms has gradually become a hot topic in recent years. Traditional tracking algorithms struggle to distinguish individual targets within the swarm under conditions of high clutter rates, low measurement rates and limited sensor resolution, resulting in limited available measurements. In this scenario, tracking each target within the swarm becomes impractical, necessitating a focus on the overall state of the UAV swarms rather than the individual states of its constituents. However, the existing methods have shown unsatisfactory tracking performance for UAV swarm in challenging scenarios with low measurement rates. To enhance the tracking performance in such challenging scenarios, in this paper, we propose a UAV swarm tracking method, called ST-UST, which combines a deep learning network named swin transformer with the Kalman filter. The core element of the proposed ST-UST method lies in the utilization of swin transformer to process images derived from noisy point clouds. In this method, swin transformer can achieve the inference of swarm shape parameters, and the Kalman filter is utilized to estimate swarm kinematic parameters. Experimental results show that, in comparison with the existing methods, the proposed ST-UST method has significant competitiveness in challenging scenarios with low measurement rates, and the maximum tracking accuracy can be improved by nearly 20.7%.