计算机科学 ›› 2024, Vol. 51 ›› Issue (6A): 230600081-7.doi: 10.11896/jsjkx.230600081
王婷, 程兰, 续欣莹, 阎高伟, 任密蜂, 张喆
WANG Ting, CHENG Lan, XU Xinying, YAN Gaowei, REN Mifeng, ZHANG Zhe
摘要： 自主定位和环境感知是机器人实现复杂任务的前提,视觉同时定位与建图(VSLAM)技术是有效解决方案。VSLAM中,传感器误差和环境噪声等影响定位和建图精度,造成累计误差。后端优化是VSLAM中消除累计误差的关键环节,现有后端优化算法通常以高斯噪声为前提,属于MSE标准下的后端算法。然而,由于图像的非凸特性和真实场景中产生的非高斯噪声,高斯噪声假设并不总成立,导致现有算法在真实场景中运行时性能下降。鉴于此,利用最大互相关熵(MCC)标准在处理非高斯噪声方面的优势,提出了一种基于MCC标准的后端优化方法,并将所提出方法应用于ORB-SLAM2框架,以测试所提出的方法在定位和建图精度方面的性能。最后,在EuRoC和KITTI公开数据集上进行实验,结果表明,无论是室内还是室外,所提方法在大部分序列中性能优于原ORB-SLAM2中基于Huber的后端优化算法以及基于Cauchy的后端优化算法。
中图分类号:
| [1]SUMIKURA S,SHIBUYA M,SAKURADA K.Openvslam:A versatile visual slam framework[C]//Proceedings of the 27th ACM International Conference on Multimedia,2019:2292-2295. [2]CONCHA A,CIVERA J.An evaluation of robust cost functions for RGB direct mapping[C]//European Conference on Mobile Robots.IEEE,2015:1-8. [3]LIU W,POKHAREL P P,PRINCIPE J C.Correntropy:A localized similarity measure[C]//The 2006 IEEE International Joint Conference on Neural Network Proceedings.IEEE,2006:4919-4924. [4]CHEN B,XING L,LIANG J,et al.Steady-state mean-square error analysis for adaptive filtering under the maximum correntropy criterion[J].IEEE signal Processing Letters,2014,21(7):880-884. [5]BARRON J T.A general and adaptive robust loss function[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019:4331-4339. [6]WILSON A M,PANIGRAHI T,MISHRA B P,et al.Adaptive Geman-McClure Estimator for Robust Distributed Channel Estimation[J].IEEE Access,2021(99):1-1. [7]PRINCIPEJ C.Information Theoretic Learning:Renyi’s Entropy and Kernel Perspectives[M].Springer Publishing Company,Incorporated,2010. [8]CHENG L,WANG K,REN M,et al.Adaptive Filter Approach for GPS Multipath Estimation Under Correntropy Criterion in Dynamic Multipath Environment[J].IEEE Transactions on Signal Processing,2019,67(22):5798-5810. [9]CHENG L,WANG K,REN M,et al.Comprehensive Analysisof Multipath Estimation Algorithms in the Framework of Information Theoretic Learning[J].IEEE Access,2018(99):1-1. [10]SUN Q N.Research on the Back-end of SLAM Based on Maximum Correntropy Criterion Filtering[D].Chongqing:Southwest University,2021. [11]LOURAKISM.SBA:A software package for generic sparsebundle adjustment[J].Acm Trans.math.softw,2009,36:1-30. [12]ARRA,Á,CHIN T J,ERIKSSON A,et al.Visual SLAM:Why Bundle Adjust?[C]//International Conference on Robotics and Automation(ICRA).IEEE,2019:2385-2391. [13]MACTAVISH K,BARFOOT T D.At all Costs:A Comparison of Robust Cost Functions for Camera Correspondence Outliers[C]//Conference on Computer & Robot Vision.IEEE ComputerSociety,2015:62-69. [14]MORÉ J J.The levenberg-marquardt algorithm:implementation and theory,in:Numerical Analysis[C]//Proceedings of the Biennial Conference Held at Dundee.Springer,2006:105-116. [15]KAUTSKY J,GOLUB G.On the calculation of jacobi matrices[J].Linear Algebra and Its Applications,1983,52:439-455. [16]MARTÍNEZ J M.Algorithms for solving nonlinear systems of equations[J].Algorithms for continuous optimization:the state of the art,1994,434:81-108. [17]BURRI M,NIKOLIC J,GOHL P,et al.The EuRoC micro aerial vehicle datasets[J].International Journal of Robotics Research,2016,35(10):1157-1163. [18]GEIGER A,LENZ P,STILLERC,et al.Vision meets robotics:The KITTI dataset[J].International Journal of Robotics Research,2013,32(11):1231-1237. [19]MUR-ARTAL R,TARDÓS J D.ORB-SLAM2:An Open-Source SLAM System for Monocular,Stereo,and RGB-D Cameras[J].IEEE Transactions on Robotics,2017,33:1255-1262. [20]KUMMERLE R,GRISETTI G,STRASDAT H,et al.G2o:A general framework for graph optimization[C]//IEEE International Conference on Robotics & Automation.IEEE,2011,3607-3613. |
|
||
首页