基于分层SLAM的空地多智能体协同导航  

Collaborative navigation of air-ground multi-agent based on hierarchical SLAM

作　　者：王晓龙 刘海颖 王景琪 WANG Xiaolong;LIU Haiying;WANG Jingqi(College of Astronautics,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)

机构地区：[1]南京航空航天大学航天学院

出　　处：《系统工程与电子技术》2020年第1期166-171,共6页Systems Engineering and Electronics

基　　金：中央高校基本科研业务费专项资金（NS2019047）;江苏省自然科学基金（BK20161490）资助课题

摘　　要：针对单一智能体在导航过程中存在全球导航卫星系统(global navigation satellite system,GNSS)易受遮挡或干扰,惯性导航存在误差累积的问题,提出基于视觉的分层即时定位与地图构建(simultaneous localization and mapping,SLAM)空地多智能体协同算法。通过建立系统模型,采用基于扩展卡尔曼滤波融合欧氏点、逆深度点、锚定同质点3种不同特征点的分层SLAM算法,实现了对导航系统的辅助和增强。针对空地协同场景设计并开展了仿真实验。结果表明,空地多智能体协同算法可以将位置误差降低40%;而在使用锚定同质点以后,误识别率由49%降低至4%。实验验证该算法具有良好的定位精度、实用性和有效性。Aiming at the problem that global navigation satellite system(GNSS)is easily occluded or interfered in the navigation process and error accumulation of inertial navigation,a vision based hierarchical simultaneous localization and mapping(SLAM)air-ground multi-agent coordination algorithm is proposed.By establishing the system model,the hierarchical SLAM algorithm based on the extended Kalman filter is used to fuse three different feature points including the Euclidean point,the inverse depth point and the anchor one,so the navigation system is assisted and enhanced.The simulation experiment is designed and carried out for the air-ground collaborative scene.The results show that the air-ground coordination method can reduce the position error down by 40%;after using the anchored homogenous point,the false recognition rate is reduced from 49%to 4%.The experimental results show that the algorithm has good positioning accuracy,practicability and effectiveness.

关 键 词：视觉导航 空地多智能体 分层即时定位与地图构建 协同导航 广义回环检测 扩展卡尔曼滤波 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置] TP391.41[自动化与计算机技术—控制科学与工程]