空中多平台网络化感知信息按需提取与融合方法研究

Network-centric information distillation and fusion is the inevitable development form of information advantage acquisition for aerial multi-platforms. Under the condition of highly dynamic network, the diverse tasks undertaken by the platforms have different demands on the environment, and with the characteristics of high speed, high mobility and complex environment, detection data are often incomplete, discontinuous, contradictory, and even conflicts and other phenomena, it is easy to cause the instability of target tracking, track reporting information uncertainty and other difficult issues. At this point, the traditional information acquisition and processing mode is difficult to support agile and efficient information processing in the multi-platform networks. .Take the dynamic tasks of aerial multi-platforms as the traction, the method of information On-demand extraction and fusion in the network is studied deeply in this project to provide support for the implementation of mission-driven information assurance, which has important scientific and practical significance to further the deepening application of aerial multi-platform networks. The main content includes: (1) Mission-oriented architecture for multi-source information fusion, which supports the on-demand target tracking; (2) Multi-source information quality assessment and system error transfer modeling, to achieve the on-demand information distillation and integration; (3) Robust track correlation and recognition verification in highly dynamic networks, to further enhance the average quality of target tracks.

网络化信息感知与处理是空中多平台信息优势获取的必然发展形态。由于网络化条件下空中多平台承担的任务日趋多样且动态变化，单一的目标信息难以满足所有平台的任务要求；且平台高速高机动条件下常常出现观测数据不连续、冲突甚至矛盾等现象，加之应用环境日益复杂，容易造成目标跟踪不稳定、目标航迹不确定等难点问题。此时，传统多源信息处理模式及方法难以支撑空中多平台网络敏捷高效的信息保障需求。.本课题以空中多平台多样化任务为牵引，深入研究网络化感知信息按需提取与融合方法，为实现任务驱动的信息按需保障能力提供支撑，对推进空中多平台网络的深化应用具有重要的科学与实际意义。主要研究内容包括：（1）面向任务的空中多平台多源信息融合架构，支撑按需定制的目标航迹生成；（2）多源信息质量评估及误差传递建模，实现网络化感知信息的按需提取与整合；（3）高动态网络化条件下的航迹稳健关联与融合识别印证，进一步提高目标航迹生成质量。

态势感知是空中多平台执行多样化任务的重要基础，信息融合是实现多源海空目标检测、跟踪、识别及威胁评估的核心环节。本项目针对高动态网络化条件下空中多平台目标信息按需保障问题，以多样化任务为牵引，建立了任务驱动的多源信息融合架构，提出了面向任务的态势信息需求模型，突破了复杂高动态环境下多源信息质量评估、高机动平台误差传递建模、目标单一航迹生成及目标综合识别印证等关键技术，可有效提升多平台网络化信息处理效率与准确性。相关研究成果在预警探测及雷达数据处理等领域得到了初步应用，获得中国航空学会科技进步二等奖1项。项目成果经转化后，在空基网络监视、天基多源协同探测、目标搜索与跟踪、重点作战区域监视与管控、海上搜索与救援等军民领域将有广泛的应用前景。.本项目实施过程中，在IEEE Transactions on Knowledge and Data Engineering, Expert Systems with Applications、Chinese Journal of Aeronautics 等国际期刊和本领域重要的国际会议(ISIF等)上发表论文26篇（均标注本项目资助），其中在SCI期刊发表论文17篇，出版论著一部《无线传感器网络监视跟踪理论与应用》；申请国家发明专利 20 项，已授权 8 项；登记软件著作权4项；培养博士研究生2名、硕士研究生3名，获全军优秀学位论文奖3人次，入选省部级以上人才计划3人次；积极参与国际及国内信息融合大会并做分会报告，与国内外同行进行了深入的学术交流与合作。.下一步，将针对天空海岸立体化探测背景，重点开展面向多样化任务需求的多平台网络化信息感知与融合理论研究，突破多源混杂信息快速获取与处理、目标信息相关性度量及按需融合等关键核心技术。