【泡泡一分钟】基于声纳的密集水下场景重建

每天一分钟，带你读遍机器人顶级会议文章

标题：Dense, Sonar-based Reconstruction of Underwater Scenes

作者：Pedro V. Teixeira, Dehann Fourie, Michael Kaess, and John J. Leonard

来源：IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)，2019

编译：柴毅

审核：黄思宇，孙钦

这是泡泡一分钟推送的第 535 篇文章，欢迎个人转发朋友圈；其他机构或自媒体如需转载，后台留言申请授权

摘要

通常，重建问题分为三个独立的步骤：首先，根据前端的要求，使用传感器处理技术来过滤和分割传感器数据。其次，前端建立问题的因子图，以获得机器人完整轨迹的精确估计。最后，通过对传感器数据的进一步处理，得到最终结果，现在从优化的轨迹重新投影。在本文中，我们提出了一种将上述问题结合在一个特定应用框架下的重建问题建模方法：基于声纳的水下结构检测。这是通过将声纳分割和点云重建问题与SLAM问题一起作为因子图来实现的。我们用船体检验试验的数据提供了试验结果。

图1 重构问题的因子图模型

图2 典型单波束回波强度测量得到的归一化经验分布。不同的曲线显示了去除最低50、90和99%的效果。

图3 数据集中有效像块的主成分分布-这些像块是平面的（𝜆2/︀𝜆≪1）和近似圆形的（𝜆0≈𝜆1）

图4 分割输出

图5 基于里程和SLAM的地图估计

Abstract

Typically, the reconstruction problem is addressed in three independent steps: first, sensor processing techniques are used to filter and segment sensor data as required by the front end. Second, the front end builds the factor graph for the problem to obtain an accurate estimate of the robot’s full trajectory. Finally, the end product is obtained by further processing of sensor data, now re-projected from the optimized trajectory. In this paper we present an approach to model the reconstruction problem in a way that unifies the aforementioned problems under a single framework for a particular application: sonar-based inspection of underwater structures. This is achieved by formulating both the sonar segmentation and point cloud reconstruction problems as factor graphs, in tandem with the SLAM problem. We provide experimental results using data from a ship hull inspection test.