多传感器融合状态估计方法综述

随着传感器网络技术的发展，多传感器融合状态估计凭借其鲁棒性、灵活性、可扩展性以及便于故障检测等优点，长期受到国内外学者的广泛关注，并取得了大量研究成果。数据融合的方法为融合状态估计奠定了理论基础，也是早期研究的主要方向，从20世纪70年代到20世纪末，相继发展出了集中式和分散式滤波架构及相应算法。无线通信技术的成熟以及一致性算法的出现使得分布式状态估计的研究进入了快车道，自2005年以来，大量基于一致性的分布式滤波算法被提出，其中不乏实用的经典方法和优秀的开创性方法。旨在梳理多传感器融合状态估计的发展，探究从数据融合到分布式滤波的内在联系，并对一些经典方法进行了总结。

A Survey of Multisensor Data Fusion State Estimation Methods

With the development of sensor network technology, multisensor state estimation has received extensive attention from scholars around the world due to its advantages of robustness, flexibility, scalability, fault detection ability. The method of data fusion lays a theoretical foundation for distributed state estimation and is also the main direction of early research. From 1970s to the end of the 20th century, centralized and decentralized filtering architectures and corresponding algorithms were successively developed. The maturity of wireless communication technology and the emergence of consensus algorithms have brought the research of distributed state estimation into the fast lane. Since 2005, a large number of distributed filtering algorithms based on consensus have been proposed, among which there are many practical classical methods and excellent pioneering methods. This paper aims to review the development of multisensor data fusion state estimation, explore the internal connection from data fusion to distributed filtering, and provide insights into the development of distributed state estimation. Some classical methods are summarized.