基于移动平台的近海渔业资源智能监测技术研究

Since the last century, China’s offshore fishery ecological environment has been severely damaged. Marine fishery resources are becoming depleted. Marine ecological environmental protection has become an important topic in the field of the national technology development. The study on intelligent monitoring technologies of fisheries resources near Zhoushan sea area based on mobile platform are performed. Several theories are discussed, including methods of fish acoustic imaging, detection, classification and recognition based on mobile platforms, the hydrodynamic environment modeling and calculation of temperature and salinity fields, and analysis of the relationship between offshore fish distribution and the hydrodynamic environment. Through the study of these theories, the theory frame of intelligent fishery environment monitoring is constructed, which will provide new methods and ideas for intelligent monitoring for the offshore fishery resource..On the technical side, three key technologies will be addressed, including the clock synchronization technology of multi-platform and multi-sensor (imaging sonar, CTD, navigation systems), method of Multi-source data assimilation, and techniques of global optimization deployment for mobile platforms. Through the solution of the above techniques, mobile platforms with intelligent monitoring for the fishery environment will be built..Through the theoretical calculation and analysis, simulations, and real sea test, the theory and technology of intelligent monitoring for China’s offshore fisheries resources and environment will be formed. The results of the project will enhance the level of marine eco-environmental protection and ensure the sustainable exploitation of fishery resources in the East China Sea.

上个世纪以来，近海渔场生态环境受到严重破坏，海洋渔业资源日趋枯竭，海洋生态环境保障已成为国家科技发展高度重视的主题。项目开展基于移动平台的舟山及附近海域渔业资源智能监测技术研究，通过解决移动平台鱼群声学成像、检测、分类和识别方法，与鱼群活动紧密关联的近海水动力环境建模及温盐场计算，以及近海鱼群分布与水动力环境耦合分析等科学问题，建立渔场环境智能监测理论框架，为近海渔业资源的智能监测提供新方法和新思路。.项目在技术方面主要解决多平台多传感（声纳、CTD、导航系统等）时钟同步技术、多源数据同化技术、移动平台全局优化部署技术等，通过相关技术的解决，构建渔场环境智能监测移动平台。.在理论及技术研究的基础上，通过仿真分析、湖上及海上试验，形成基于移动平台的近海渔业资源及渔场环境智能监测理论与技术体系的能力，项目的成果对提升我国海洋生态环境保护水平，保障东海渔业资源的可持续开发具有重要的战略意义。

海洋生态环境保障已成为国家科技发展高度关注的领域，本项目开展基于移动平台的舟山及附近海域渔业资源智能监测技术研究，根据项目计划书，经过四年的研究，项目取得的成果包括：1）完成了自主水下航行器平台（Autonomous Underwater Vehicle，AUV）的改装以及鱼群成像声呐、水下摄像头、CTD等载荷的集成和软件开发，研制完成具有鱼群声光智能监测功能的AUV舱段。2）针对单个机器学习算法在鱼群分类方面的局限性，为进一步提高基于声学回波数据的鱼种分类结果的准确率，提出了一种基于多模型改进的Stacking集成学习鱼类识别方法，利用仿真及实验验证了该方法的有效性。3）搭建了东中国海动力数值模型并进行验证，在此基础上，开展了舟山渔场叶绿素反演及浮游生物分布分析。4）在海洋动力模式温度场预报的基础上，开展了移动平台海洋环境观测全局路径规划算法研究，完成算法实现，并验证了其有效性。5）开展了多次湖上及海上试验，验证了基于移动平台的鱼群智能监测系统各项功能的有效性，为该系统后续在实际中的推广应用奠定了基础。