没有合适的资源?快使用搜索试试~ 我知道了~
基于学习方法的高精度SLAM算法研究1
需积分: 0 0 下载量 15 浏览量
2022-08-03
20:42:35
上传
评论
收藏 2.32MB PDF 举报
温馨提示
试读
68页
绪论1.1 课题背景及研究意义
资源详情
资源评论
资源推荐
硕硕硕士士士学学学位位位论论论文文文
基于学习方法的高精度SLAM算法研究
RESEARCH ON LEARNING BASED
HIGH PRECISION SLAM ALGORITHM
冯冯冯爱爱爱迪迪迪
哈哈哈尔尔尔滨滨滨工工工业业业大大大学学学
2018 年年年 6 月月月
国内图书分类号:TP391.4
国际图书分类号:621.3
学校代码:10213
密级:公开
工工工学学学硕硕硕士士士学学学位位位论论论文文文
基于学习方法的高精度SLAM算法研究
硕 士 研 究 生:冯爱迪
导 师:张大鹏教授
申 请 学 位:工学硕士
学 科:计算机科学与技术
所 在 单 位:计算机科学与技术学院
答 辩 日 期:2018 年 6 月
授予学位单位:哈尔滨工业大学
Classified Index: TP391.4
U.D.C: 621.3
Dissertation for the Master’s Degree in Engineering
RESEARCH ON LEARNING BASED
HIGH PRECISION SLAM ALGORITHM
Candidate: Aidi Feng
Supervisor: Prof. David Zhang
Academic Degree Applied for: Master of Engineering
Specialty: Computer Science and Technology
Affiliation: School of Computer Science and Technology
Date of Defence: June, 2018
Degree-Conferring-Institution: Harbin Institute of Technology
哈尔滨工业大学工学硕士学位论文
摘 要
同时定位和地图构建(Simultaneous Localization and Mapping,SLAM)是一
种三维重建的方法。它是一种在未知环境中自主定位并进行地图构建的方法。
SLAM在自动驾驶、虚拟现实等多个领域都有广泛的应用。
在众多SLAM算法当中,LSD-SLAM算法是一种可以实时地对大规模场景
进行重建的方法。但是LSD-SLAM算法还存在提升的空间。近些年,基于学习
的方法活跃在各个领域中,并取得了不错的成绩。因此,本文尝试利用基于学
习的方法提升LSD-SLAM算法的重建鲁棒性。
SLAM系统由传感器、前端视觉里程计、后端非线性优化、闭环检测和地
图构建五部分组成。其中,视觉里程计部分和闭环检测部分可以加入基于学习
的方法,本文从这两个方面对LSD-SLAM算法做了改进。首先本文提出了一个
基于可信控制点的深度置信度估计算法,对LSD-SLAM的视觉里程计部分做了
提升和改进。置信度估计算法使用随机森林算法训练可信控制点预测模型,模
型使用在立体匹配过程中可以方便快捷的计算出的特征,使模型在保证准确度
的情况下,时间的消耗尽量小。在跟踪估计的过程中,通过模型得到一个深度
置信度的估计,将置信度作为权值融入到深度估计和相机运动估计当中,得到
更为准确的估计结果,提升LSD-SLAM算法在前端视觉里程计部分的精度,进
而提升整个系统的重建精度。
本文还提出了一种基于二阶特征的闭环检测网络模型。模型采用深度卷积
网络,基于二阶的特征信息,得到高精度的闭环检测结果。模型采用的损失函
数为三元组损失函数。通过这种弱监督学习,不断缩小同一个地点的特征之间
的距离,不断加大不同地点的特征距离,使得相同地点的特征聚为一类。将提
出的闭环检测网络模型放入LSD-SLAM当中,提升模型在闭环检测环节的准确
性,使重建的结果更加的准确。
利用上述两个模型,本文对LSD-SLAM算法的重建精度和鲁棒性做了提升,
使算法可以得到更好的重建效果。
关键词:三维重建;可信控制点;立体匹配;闭环检测;二阶特征
- I -
哈尔滨工业大学工学硕士学位论文
Abstract
Simultaneous Localization and Mapping (SLAM) is a 3D reconstruction method. It
is a method of self-localization and mapping in an unknown environment. SLAM has a
wide range of applications in areas such as automatic driving and virtual reality.
Among various SLAM algorithms, the LSD-SLAM algorithm is a method that can
reconstruct large-scale scenes in real time. However, there is room for improvement in
the LSD-SLAM algorithm. In recent years, learning-based methods have been active in
various fields and have achieved good results. Therefore, this project attempts to use
learning-based methods to improve the robustness of LSD-SLAM.
The SLAM system consists of a sensor, a front-stage visual odometry, back-stage
nonlinear optimization, loop closure, and mapping. Among them, the visual odometry
part and the loop closure part can join the learning-based method. This subject has im-
proved the LSD-SLAM algorithm from these two aspects. Firstly, this topic proposes a
depth confidence estimation algorithm based on ground control points, which improves
the visual odometry part of the LSD-SLAM. The confidence estimation algorithm uses a
random forest algorithm to train the gorund control points prediction model. The mod-
el can use the features that can be calculated conveniently and quickly in the process of
stereo matching, so that the model costs as little as possible while prediction. In the pro-
cess of tracking estimation, an estimate of the depth confidence is obtained through the
model, and the confidence is used as a weight into the depth estimation and the camer-
a motion estimation to obtain a more accurate estimation result and improve the LSD-
SLAM algorithm in the front-stage visual odometry part accuracy, and then improve the
reconstruction accuracy of the entire system.
This topic also proposes a loop closure detection network model based on second-
order features. The model uses a deep convolutional network, based on the second-order
information, to obtain high-precision loop closure detection results. The loss function
used by the model is the triplet loss function. Through this kind of weakly supervised
learning, the distance between the features of the same place is continuously reduced, and
the feature distances of different locations are constantly increased, so that the features of
the same place keep clustering. The proposed loop closure detection network model is put
- II -
剩余67页未读,继续阅读
文润观书
- 粉丝: 26
- 资源: 317
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功
评论0