泡泡一分钟:Fast and Robust Initialization for Visual-Inertial SLAM
张宁 Fast and Robust Initialization for Visual-Inertial SLAM
链接:https://pan.baidu.com/s/1cdkuHdkSi9x7l-96zMbX7g 提取码:b3ff
Carlos Campos, Jos´e M.M. Montiel and Juan D. Tard´os
Visual-inertial SLAM (VI-SLAM) requires a good initial estimation of the initial velocity, orientation with respect to gravity and gyroscope and accelerometer biases.In this paper we build on the initialization method proposed by Martinelli [1] and extended by Kaiser et al. [2], modifying it to be more general and efficient. We improve accuracy with several rounds of visual-inertial bundle adjustment, and robustify the method with novel observability and consensus tests, that discard erroneous solutions. Our results on the EuRoC dataset show that, while the original method produces scale errors up to 156%, our method is able to consistently initialize in less than two seconds with scale errors around 5%, which can be further reduced to less than 1% performing visual-inertial bundle adjustment after ten seconds.
视觉惯性SLAM(VI-SLAM)需要对初始速度,相对于重力和陀螺仪的方向以及加速度计偏差进行良好的初始估计。在本文中,我们建立了Martinelli提出的初始化方法[1]并由Kaiser等人[2]扩展,将其修改为更一般和更有效。我们通过几轮视觉惯性束调整来提高准确性,并通过新颖的可观察性和共识测试来证明该方法,从而丢弃错误的解决方案。我们在EuRoC数据集上的结果表明,虽然原始方法产生的标度误差高达156%,但我们的方法能够在不到两秒的时间内始终如一地进行初始化,标度误差约为5%,视觉惯性束调整进行十秒钟后可进一步降低至小于1%。
泡泡一分钟:Fast and Robust Initialization for Visual-Inertial SLAM的更多相关文章
- 泡泡一分钟:Robust and Fast 3D Scan Alignment Using Mutual Information
Robust and Fast 3D Scan Alignment Using Mutual Information 使用互信息进行稳健快速的三维扫描对准 https://arxiv.org/pdf/ ...
- 泡泡一分钟:Robust Attitude Estimation Using an Adaptive Unscented Kalman Filter
张宁 Robust Attitude Estimation Using an Adaptive Unscented Kalman Filter 使用自适应无味卡尔曼滤波器进行姿态估计链接:https: ...
- 泡泡一分钟:FMD Stereo SLAM: Fusing MVG and Direct Formulation Towards Accurate and Fast Stereo SLAM
FMD Stereo SLAM: Fusing MVG and Direct Formulation Towards Accurate and Fast Stereo SLAM FMD Stereo ...
- 泡泡一分钟: A Linear Least Square Initialization Method for 3D Pose Graph Optimization Problem
张宁 A Linear Least Square Initialization Method for 3D Pose Graph Optimization Problem "链接:https ...
- 泡泡一分钟:LandmarkBoost: Efficient Visual Context Classifiers for Robust Localization
Marcin Dymczyk, Igor Gilitschenski, Juan Nieto, Simon Lynen, Bernhard Zeis, and Roland Siegwart Land ...
- 泡泡一分钟:Efficient Trajectory Planning for High Speed Flight in Unknown Environments
张宁 Efficient Trajectory Planning for High Speed Flight in Unknown Environments 高效飞行在未知环境中的有效轨迹规划链接: ...
- 泡泡一分钟:Exploiting Points and Lines in Regression Forests for RGB-D Camera Relocalization
Exploiting Points and Lines in Regression Forests for RGB-D Camera Relocalization 利用回归森林中的点和线进行RGB-D ...
- 泡泡一分钟:A Multi-Position Joint Particle Filtering Method for Vehicle Localization in Urban Area
A Multi-Position Joint Particle Filtering Method for Vehicle Localization in Urban Area 城市车辆定位的多位置联合 ...
- 泡泡一分钟:Tightly-Coupled Aided Inertial Navigation with Point and Plane Features
Tightly-Coupled Aided Inertial Navigation with Point and Plane Features 具有点和平面特征的紧密耦合辅助惯性导航 Yulin Ya ...
随机推荐
- Caused by: java.nio.charset.MalformedInputException: Input length = 1
java.lang.IllegalStateException: Failed to load property source from location 'classpath:/applicatio ...
- 常用的HTTP状态码,网站开发请求状态必备
成功的状态码: 200 – 服务器成功返回网页 304 – 未修改 失败的状态码: 404 – 请求的网页不存在 503 – 服务器暂时不可用 500 – 服务器内部错误 下面的不是很常用,记住上面那 ...
- 什么是ARP协议?
ARP协议,全称“Address Resolution Protocol”,中文名是地址解析协议, 使用ARP协议可实现通过IP地址获得对应主机的物理地址(MAC地址). 在TCP/IP的网络环境下, ...
- Kafka消息丢失
1.Kafka消息丢失的情况: (1)auto.commit.enable=true,消费端自动提交offersets设置为true,当消费者拉到消息之后,还没有处理完 commit interval ...
- 洛谷 P2279 [HNOI2003]消防局的设立 题解
每日一题 day34 打卡 Analysis 这道题的正解本来是树形dp,但要设5个状态,太麻烦了.于是我就用贪心试图做出此题,没想到还真做出来了. 考虑当前深度最大的叶子结点,你肯定要有一个消防局去 ...
- CDH 大数据平台搭建
一.概述 Cloudera版本(Cloudera’s Distribution Including Apache Hadoop,简称“CDH”),基于Web的用户界面,支持大多数Hadoop组件,包括 ...
- Intel 80386 CPU
一.80386 概述 80386处理器被广泛应用在1980年代中期到1990年代中期的IBM PC相容机中.这些PC机称为「80386电脑」或「386电脑」,有时也简称「80386」或「386」.80 ...
- php技能树---大神的进阶之路
PHP7 迎来巨大的性能提升,又一次回到关注的焦点.根据这些年在开发圈子总结的LNMP程序猿发展轨迹,结合个人经验体会,总结出很多程序员对未来的迷漫,特别对技术学习的盲目和慌乱,简单梳理了这个每个阶段 ...
- fft相关的复习
任意长度卷积 CZT 就是一波推导 \[ \begin{aligned} b_i &= \sum_{j=0}^{n-1} \omega^{ij}a_j \\ &= \sum_{j=0} ...
- (11)打鸡儿教你Vue.js
表单 v-model 指令在表单控件元素上创建双向数据绑定 <div id="app"> <p>单个复选框:</p> <input typ ...