当前位置: X-MOL 学术Auton. Robot. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
RGB-D camera calibration and trajectory estimation for indoor mapping
Autonomous Robots ( IF 3.7 ) Pub Date : 2020-08-17 , DOI: 10.1007/s10514-020-09941-w
Liang Yang , Ivan Dryanovski , Roberto G. Valenti , George Wolberg , Jizhong Xiao

In this paper, we present a system for estimating the trajectory of a moving RGB-D camera with applications to building maps of large indoor environments. Unlike the current most researches, we propose a ‘feature model’ based RGB-D visual odometry system for a computationally-constrained mobile platform, where the ‘feature model’ is persistent and dynamically updated from new observations using a Kalman filter. In this paper, we firstly propose a mixture of Gaussians model for the depth random noise estimation, which is used to describe the spatial uncertainty of the feature point cloud. Besides, we also introduce a general depth calibration method to remove systematic errors in the depth readings of the RGB-D camera. We provide comprehensive theoretical and experimental analysis to demonstrate that our model based iterative-closest-point (ICP) algorithm can achieve much higher localization accuracy compared to the conventional ICP. The visual odometry runs at frequencies of 30 Hz or higher, on VGA images, in a single thread on a desktop CPU with no GPU acceleration required. Finally, we examine the problem of place recognition from RGB-D images, in order to form a pose-graph SLAM approach to refining the trajectory and closing loops. We evaluate the effectiveness of the system on using publicly available datasets with ground-truth data. The entire system is available for free and open-source online.



中文翻译:

室内制图的RGB-D摄像机校准和轨迹估计

在本文中,我们提出了一种用于估计移动RGB-D相机轨迹的系统,该系统可用于构建大型室内环境的地图。与当前大多数研究不同,我们为计算受限的移动平台提出了一种基于“特征模型”的RGB-D视觉里程表系统,其中“特征模型”是持久的,并使用卡尔曼滤波器从新观测值中动态更新。本文首先提出一种混合高斯模型进行深度随机噪声估计,该模型用于描述特征点云的空间不确定性。此外,我们还介绍了一种通用的深度校准方法,以消除RGB-D相机的深度读数中的系统误差。我们提供了全面的理论和实验分析,以证明我们的基于模型的迭代最近点(ICP)算法与常规ICP相比可以实现更高的定位精度。视觉测距法在VGA图像上以30 Hz或更高的频率在台式机CPU上的单线程中运行,而无需GPU加速。最后,我们研究了从RGB-D图像进行位置识别的问题,以便形成一种姿势图SLAM方法来细化轨迹和闭合环。我们评估使用公开数据集和真实数据的系统的有效性。整个系统可免费在线免费获得。在台式机CPU上的单线程中运行,不需要GPU加速。最后,我们研究了从RGB-D图像进行位置识别的问题,以便形成一种姿势图SLAM方法来细化轨迹和闭合环。我们评估使用公开数据集和真实数据的系统的有效性。整个系统可免费在线免费获得。在台式机CPU上的单线程中运行,不需要GPU加速。最后,我们研究了从RGB-D图像进行位置识别的问题,以便形成一种姿势图SLAM方法来细化轨迹和闭合环。我们评估使用公开数据集和真实数据的系统的有效性。整个系统可免费在线免费获得。

更新日期:2020-08-17
down
wechat
bug