| 摘 要: 由于SLAM(Simultaneous Localization and Mapping)算法能够在陌生环境中进行高精准度的实时定位 以及对当前环境进行重建地图的特点,SLAM技术逐渐成为当前研究热点。本文主要研究基于图优化的同时定位与地图 创建,即SLAM创建中非线性图优化的算法。在基于图优化的SALM问题中,最主要的就是解决非线性最小二乘问题。 本文对非线性最小二乘问题的算法和常见的非线性优化方案进行阐述与分析,分析最速下降法、高斯—牛顿法、列文伯 格—马夸尔特法的原理和步骤,总结比较三种方法的特征和缺点,在SLAM框架中选择最适合的优化算法。 |
| 关键词: 图优化;非线性最小二乘;SLAM |
|
中图分类号: TP391.41
文献标识码: A
|
|
| Research on Simultaneous Localization and Mapping based on Nonlinear Optimization Algorithm |
|
JI Chen, SONG Yanyan, QIN Jun, AO Tiantian
|
(Communication University of China, Nanjing, Nanjing 211172, China)
592619059@qq.com; sophiesong1231@163.com; 1059182465@qq.com; 986952863@qq.com
|
| Abstract: This paper studies nonlinear graph optimization algorithm in Simultaneous Localization and Mapping (SLAM). In the SLAM problem, the major issue is to solve the nonlinear least squares problem. In this paper, algorithm of nonlinear least squares problem and common nonlinear optimization schemes are analyzed, such as the principles and steps of the steepest descent method, Gauss Newton method and Levenberg-Marquardt method. Characteristics and shortcomings of the three methods are compared with each other, and the most suitable optimization algorithm in the SLAM framework is selected. |
| Keywords: graph optimization; nonlinear least squares; SLAM |
