首页|基于改进关键帧筛选的多状态约束卡尔曼滤波

基于改进关键帧筛选的多状态约束卡尔曼滤波

扫码查看
基于多状态约束卡尔曼滤波的融合算法仅利用单帧图像进行位姿估计,若初始化不正确,会导致视觉位姿估计发散严重;若将每个视觉特征点都包含进系统状态向量,则极易增加系统计算负担.针对上述问题,提出了一种改进的关键帧选择算法,利用多个视觉关键帧对相同特征点的约束来减小视觉测量误差,提高定位精度,同时只将关键帧解算出的相机位姿融入系统状态向量,有效地降低了系统计算量.实验表明,改进算法与EKF相比,其定位精度和计算效率分别提升了29.09%和32.2%.与Orb-slam2相比,改进算法的计算效率提升了35.48%.
Multi-State Constrained Kalman Filtering Based on Improved Keyframe Filtering
The fusion algorithm based on multi-state constrained Kalman filtering solely uses a single frame im-age for pose estimation.If the initialization is incorrect,then it can cause severe divergence in visual pose estima-tion.Furthermore,each visual feature point in the system state vector can easily lead to computational burden to the system.Given the aforementioned problems,an improved keyframe selection algorithm is proposed,which uses multiple visual keyframes to constrain the same feature points for reducing visual measurement errors and improving positioning accuracy.Simultaneously,only the camera pose calculated from keyframes is integrated into the system state vector,which can effectively reduce system computation.The experiment shows that the improved algorithm enhances po-sitioning accuracy and computational efficiency by 29.09%and 32.2%,respectively,when compared to EKF.Additional-ly,the proposed algorithm increases computational efficiency by 35.48%when compared to that of Orb-slam2.

keyframesKalman filterfusion algorithmvisual positioningINS

修瑾智、方针、彭慧、陈燕苹、邹梦强、刘宇、杨诚霖、王森

展开 >

重庆邮电大学 自主导航与微系统重庆市重点实验室,重庆 400065

重庆邮电大学 智能传感技术与微系统重庆市高校工程研究中心,重庆 400065

中国电子科技集团公司 第二十六研究所,重庆 400060

关键帧 卡尔曼滤波 融合算法 视觉定位 惯性定位

国家自然科学基金资助项目国家自然科学基金资助项目重庆市自然科学基金项目资助重庆市自然科学基金项目资助重庆市自然科学基金项目资助重庆市自然科学基金项目资助

5217553162305039CSTB2023NSCQ-MSX0568CSTB2022NSCQ-LZX0050CSTB2023NSCQ-LMX0028cstc2022ycjhbgzxm0190

2024

压电与声光
四川压电与声光技术研究所

压电与声光

CSTPCD北大核心
影响因子:0.357
ISSN:1004-2474
年,卷(期):2024.46(4)