基于双目视觉的绳驱动飞行机械臂目标识别与抓取  

Target Tracking and Grabbing of Cable-driven Aerial Manipulator Based on Binocular Vision

作　　者：谭荣凯 杨浩秦 王尧尧 TAN Rongkai;YANG Haoqin;WANG Yaoyao(College of Mechanical and Electrical Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China;College of Materials Science and Technology,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)

机构地区：[1]南京航空航天大学机电学院,江苏南京210016 [2]南京航空航天大学材料科学与技术学院,江苏南京210016

出　　处：《机械制造与自动化》2024年第1期191-195,共5页Machine Building & Automation

基　　金：国家自然科学基金项目(52175097);浙江大学流体动力与机电系统开放基金项目(GZKF-201915)。

摘　　要：为提高面向飞行机械臂平台的目标姿态估计的精度,为飞行机械臂配备双目视觉系统,对绳驱动机械臂和双目相机进行系统建模,利用Camshift算法对目标进行实时追踪,结合目标的几何特征提出一种轻量化的目标姿态估计算法,可用于飞行平台的实时目标追踪与姿态估计并进行飞行机械臂抓取实验。实验证明:该姿态估计算法对于目标追踪和姿态估计具有较好精度,可以实现抓取作业。To improve the accuracy of target pose estimation on aerial manipulator platform,the aerial manipulator is equipped with a binocular vision system,and the cable-driven manipulator and the binocular camera are modeled.By the Camshift algorithm and with the geometric characteristics of the target,a lightweight target pose estimation algorithm is proposed,which can be used for real-time target tracking and pose estimating on aerial platform as well as for grasping experiment of aerial manipulator.Experiments show that the proposed pose estimation algorithm has good accuracy for target tracking and pose estimating,and can undertake grasping operation.

关 键 词：绳驱动飞行机械臂 双目相机 CAMSHIFT 目标追踪 姿态估计 

分 类 号：TP249[自动化与计算机技术—检测技术与自动化装置]