汽车工程 ›› 2023, Vol. 45 ›› Issue (8): 1457-1467.doi: 10.19562/j.chinasae.qcgc.2023.08.016

所属专题: 智能网联汽车技术专题-感知&HMI&测评2023年

• • 上一篇    下一篇

基于雅克比域零空间边缘化的视觉SLAM

芦涛1,金馨2,廖毅霏1,黄圣杰1,杨依琳1,谢国涛1,2,秦晓辉1,3()   

  1. 1.汽车车身先进设计制造国家重点实验室,湖南大学机械与运载工程学院,长沙  410082
    2.山东省科学技术情报研究院,济南  250101
    3.湖南大学无锡智能控制研究院,无锡  214115
  • 收稿日期:2022-12-08 修回日期:2023-02-28 出版日期:2023-08-25 发布日期:2023-08-17
  • 通讯作者: 秦晓辉 E-mail:qxh880507@163.com
  • 基金资助:
    国家自然科学基金(52102456);长沙市自然科学基金(kq2202162);汽车车身先进设计制造国家重点实验室开放课题(32115013)

Visual SLAM Based on Jacobian Null-space Marginalization

Tao Lu1,Xin Jin2,Yifei Liao1,Shengjie Huang1,Yilin Yang1,Guotao Xie1,2,Xiaohui Qin1,3()   

  1. 1.State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body,College of Mechanical and Vehicle Engineering,Hunan University,Changsha  410082
    2.Shandong Institute of Scientific and Technical Information,Jinan  250101
    3.Wuxi Intelligent Control Research Institute of Hunan University,Wuxi  214115
  • Received:2022-12-08 Revised:2023-02-28 Online:2023-08-25 Published:2023-08-17
  • Contact: Xiaohui Qin E-mail:qxh880507@163.com

摘要:

为降低系统求解大规模线性方程时的计算资源占用率、提高系统运行速度,现有基于非线性优化的视觉SLAM框架大多利用增量方程中海森矩阵稀疏性与边缘化策略对问题进行降阶。然而,这些方法仍须占用大量内存以显式构建超高维度的海森矩阵,且由于该方法对数值变化的敏感性,在实际部署时常依赖双精度浮点数进行求解以降低数值误差,限制了其在低算力平台中的应用。为解决这一问题,本文提出基于雅克比域零空间边缘化的视觉SLAM方法,该方法在后端优化模块中将路标雅克比矩阵投影至其左零空间,在避免构建海森矩阵前提下达到降阶效果,提升求解效率,并在代数上证明两种边缘化方法的等价性。从数值分析角度证明本文提出的边缘化方法具备更好的数值稳定性,可支持单精度浮点数求解,进一步提升效率。公开数据集和实车试验表明本文方法相较于基于舒尔消元边缘化的通用优化器具备更好的求解速度和精度。

关键词: 自动驾驶, 视觉SLAM, 光束法平差, 边缘化

Abstract:

The existing visual SLAM frameworks based on nonlinear optimization, in order to reduce the computational resource occupation and improve the system operation speed when solving large-scale linear equations, mostly use the Hession matrix sparsity and marginalization strategy in incremental equations to reduce the order of the problem. However, these methods still need a large amount of memory to explicitly construct the ultra-high dimensional Hessian matrix. Moreover, due to the sensitivity of the method to numerical changes, in order to reduce numerical errors, in actual deployment, they often rely on double precision floating point numbers to solve, which limits the application in low computing power platforms. To solve this problem, this paper proposes a visual SLAM method based on Jacobian domain null-space marginalization, which projects the landmark Jacobian matrix to its left null-space in the back-end optimization module, achieves the effect of reduction and improves the solution efficiency on the premise of avoiding the construction of Hessian matrix, and proves the equivalence of the two marginalization methods algebraically. From the perspective of numerical analysis, it is proved that the marginalization method proposed in this paper has better numerical stability and can support the single precision floating point solution, with further improvement of the efficiency. The open dataset and real vehicle test show that the method in this paper has better solution speed and accuracy than the general optimizer based on Schur complement marginalization.

Key words: automatic driving, visual SLAM, bundle adjustment, marginalization