一种基于多视觉传感器的刚体位姿估计方法  

A kind of pose estimation method of rigid body based on multiple visual sensors

作　　者：张文安[1] 陈玉寅 尹姝[1] 仇翔[1] ZHANG Wenan;CHEN Yuyin;YIN Shu;QIU Xiang(College of Information Engineering,Zhejiang University of Technology,Hangzhou 310023,China)

机构地区：[1]浙江工业大学信息工程学院,浙江杭州310023

出　　处：《浙江工业大学学报》2018年第6期598-604,共7页Journal of Zhejiang University of Technology

基　　金：浙江省公益技术应用研究计划项目(2016C31065);浙江省杰出青年科学基金资助项目(LR16F030005);国家自然科学基金资助项目(61573319)

摘　　要：研究了基于多视觉传感器的移动刚体位置和姿态(位姿)估计问题,提出了一种基于自适应扩展卡尔曼滤波器(AEKF)的分布式融合估计方法以提升多视觉刚体位姿估计系统的运行效率。首先,基于多视觉传感器实验平台建立了系统的观测模型和刚体的运动学模型,继而将自适应扩展卡尔曼滤波器作为局部滤波器,提出了应用于多视觉位姿跟踪系统的分布式融合方法,提高了系统的运行效率。最后,通过实验验证了所提出方法的有效性和优越性。This paper studies the moving rigid body position and posture(pose)estimation based multiple visual sensors,and proposes a kind of distributed fusion method based adaptive extended Kalman filter(AEKF)to improve operation efficiency of the rigid body estimation system based multiple visual sensors.First of all,the system of observation model and kinematics model of rigid body have been established based the visual sensors experiment platform.To improve the operation efficiency of the system,a distributed fusion method applied in multi-visual tracking system is put forward which uses adaptive extended Kalman filter as a local filter.Finally,the effectiveness and superiority of the proposed method is verified by experiments.

关 键 词：视觉 多传感器融合 非线性卡尔曼滤波器 

分 类 号：TN713[电子电信—电路与系统] TP212[自动化与计算机技术—检测技术与自动化装置]