基于相似结构约束和自适应调整的双目视觉里程计  

Binocular Visual Odometry Based on Similar Structural Constraints and Adaptive Adjustment

作　　者：张硕骁 张涛[1,2] Zhang Shuoxiao;Zhang Tao(School of Instrument Science and Engineering,Southeast University,Nanjing 210096,China;Key Laboratory of Micro-inertial Instrument and Advanced Navigation Technology of Ministry of Education,Southeast University,Nanjing 210096,China)

机构地区：[1]东南大学仪器科学与工程学院,南京210096 [2]东南大学微惯性仪表与先进导航技术教育部重点实验室,南京210096

出　　处：《战术导弹技术》2021年第2期39-47,共9页Tactical Missile Technology

基　　金：国家自然科学基金(52071080);惯性技术重点实验室基金(614250607011709);水下信息与控制重点实验室基金(614221805051809);中央高校基本科研业务费(2242020k1G009);江苏绿色船舶技术重点实验室基金(2019Z01);东南大学国家自然科学基金结题项目剩余经费培育项目(9S20172204)。

摘　　要：为提高在不同环境下双目视觉里程计的精度,提出一种基于相似结构性约束和自适应调整策略的位姿估计方法。利用特征点对的相似结构性约束对传统的RANSAC算法进行改进,降低误匹配率的同时提高匹配效率;提出一种自适应调整策略,通过跟踪特征的存活率动态调整特征匹配的质量控制阈值,在被跟踪的特征点的数量和质量之间保持一个良好的平衡并提升位姿估计的精度。最后利用KITTI数据集进行了仿真实验,结果表明提出算法相对于传统的ORB-SLAM2算法,定位精度提升了约21%,满足实时定位的要求。In order to improve the accuracy of binocular visual odometry in different environments,a pose estimation method based on similar structural constraints and adaptive adjustment strategies is proposed.The traditional RANSAC algorithm is improved by using the similar structural constraints of feature point pairs to reduce the mismatching rate and improve the matching efficiency.An adaptive adjustment strategy is proposed,which dynamically adjusts the quality control threshold of feature matching through the survival rate of tracking features,so as to maintain a good balance between the quantity and quality of tracked feature points and improve the accuracy of pose estimation.Finally,KITTI data set is used for simulation experiments,the results show that the algorithm,compared to the traditional ORB-SLAM2 algorithm,improves the positioning accuracy by about 21%,which meets the requirements of real-time positioning.

关 键 词：双目视觉里程计 相似性约束 误匹配消除 自适应调整 位姿估计 

分 类 号：U666.1[交通运输工程—船舶及航道工程]