军用无人系统协同感知与分布式滤波技术  

Collaborative Perception and Distributed Filtering Technology for Military Unmanned Systems

作　　者：李诗航 孙玉洁 崔建峰[1,2] 曾建潮 卢志刚[2] LI Shihang;SUN Yujie;CUI Jianfeng;ZENG Jianchao;LU Zhigang(North University of China,Taiyuan 030051,China;North Automatic Control Technology Institute,Taiyuan 030006,China)

机构地区：[1]中北大学,太原030051 [2]北方自动控制技术研究所,太原030006

出　　处：《火力与指挥控制》2024年第3期88-94,共7页Fire Control & Command Control

基　　金：国家自然科学基金(61903343);装备预先研究资助项目(315197203)。

摘　　要：现代战争由机械化向无人化转型,而人工智能、大数据等技术的发展促使未来战争形态呈现出仿生智能的特点。仿生无人机、智能无人车等载荷打击平台通过卫星数据链开展典型区域联合作战的前提是已知敌我双方的准确位置信息。围绕军用无人运动平台的演化历程和发展趋势,对无人系统的构成和特点进行归纳,进而指出多传感器协同感知是信息获取的主要形式,通过分布式滤波处理得到平台的位置和姿态。对多平台协同感知能力、协作估计策略和SLAM技术进行分析和总结,指出融合深度智能的协同认知、数据驱动和多源多模态融合等技术是提升无人作战效能的核心关键。Modern warfare is transforming from mechanization to unmanned,and the development of artificial intelligence,big data and other technologies makes the future form of warfare inevitably show the characteristics of bionic intelligence.The premise of carrying out typical regional joint operations through satellite data links for payload strike platforms such as bionic unmanned aerial vehicles and intelligent unmanned vehicles is to know the accurate location information of both ourselves and the enemy.Focusing on the evolution history and development trend of the military unmanned motion platform,the composition and characteristics of the unmanned system are summarized,and then it is pointed out that multi-sensor collaborative perception is the main form of information acquisition,and the position and attitude of the platform are obtained through distributed filtering processing.The multi-platform collaborative perception capability,collaborative estimation strategy and SLAM technology are analyzed and summarized,and such technologies as collaborative cognition integrating deep intelligence,data-driven and multi-source multi-modal fusion,etc.are the core keys to improve the effectiveness of unmanned combat.

关 键 词：无人平台 协同感知 分布式滤波 智能认知 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]