A System for Volumetric Robotic Mapping of Abandoned Mines

A System for Volumetric Robotic Mapping of Abandoned Mines

volumetric

英 [,vɒljʊ’metrɪk] 美 [,vɑljə’mɛtrɪk]

adj. [物] 体积的；[物] 容积的；[物] 测定体积的

Abstract— This paper describes two robotic systems developed for acquiring accurate volumetric maps of underground mines. One system is based on a cart instrumented by laser range finders, pushed through a mine by people. Another is a remotely controlled mobile robot equipped with laser range finders. To build consistent maps of large mines with many cycles, we describe an algorithm for estimating global correspondences and aligning robot paths.

摘要:本文介绍了两种用于获取地下矿山精确体积图的机器人系统。其中一个系统是基于一个由激光测距仪
驱动的手推车，由人推入矿井。另一种是配备激光测距仪的遥控移动机器人。为了构建具有多个周期的大型
矿井的一致映射，我们描述了一种估计全局通信和对齐机器人路径的算法。

This algorithm enables us to recover consistent maps several hundreds of meters in
diameter, without odometric information. We report results obtained in two mines, a research mine in Bruceton, PA, and an abandoned coal mine in Burgettstown, PA.

这种算法使我们能够恢复几百米直径的一致地图，而不需要测量里程表信息。我们报告了在两个矿井中获得
的结果，一个是在宾夕法尼亚州布鲁瑟顿的一个研究矿井，另一个是在宾夕法尼亚州伯盖特镇的一个废弃煤矿。

I. INTRODUCTION
The lack of accurate maps of inactive, underground mines poses a serious threat to public safety. According to a recent article [3], “tens of thousands, perhaps even hundreds of thousands, of abandoned mines exist today in the United States. Not even the U.S. Bureau of Mines knows the exact number, because federal recording of mining claims was not required until 1976.” The lack of accurate mine maps frequently causes accidents, such as a recent near-fatal accident in Quecreek, PA [18]. Even when accurate maps exist, they provide information only in 2-D, which is usually insufficient to assess the structural soundness of abandoned mines.

不活跃地下矿井的精确地图的缺失对公共安全构成严重威胁。根据[3]最近的一篇文章，“今天在美国存在数	
万，甚至数十万废弃的地雷。甚至美国矿业局也不知道确切的数字，因为联邦政府直到1976年才要求对矿业
索赔进行记录。“缺乏精确的矿井地图经常会导致事故，比如最近在巴布0州Quecreek发生的一次近乎致命的
事故。即使有精确的地图，它们也只能提供二维的资料，这通常不足以评价废弃矿井的结构健全程度。

Hazardous operating conditions and difficult access routes suggest that robotic exploration and mapping of abandoned mines may be a viable option. The idea of mapping mines with robots is not new.

危险的操作条件和困难的通道表明，机器人勘探和绘制废弃矿井的地图可能是可行的选择。用机器人测绘矿
井的想法并不新鲜。

Past research has predominantly focused on acquiring maps for autonomous robot navigation in active mines. For example, Corke and colleagues [8] have built vehicles that acquire and utilize accurate 2-D maps of mines. Similarly, Baily [1] reports 2-D mapping results of an underground area using advanced mapping techniques.

过去的研究主要集中在获取主动水雷中自动机器人导航的地图。例如，科克和他的同事[8]已经制造了能够获
取和利用精确的地雷二维地图的车辆。与此类似，Baily[1]报告了一个地下区域的二维测绘结果，使用了先进
的测绘技术。

None of these techniques generate volumetric maps of mines.

In general, the mine mapping problem is made challenging by the lack of global position information underground.As a result, mine mapping must be approached as a simultaneous localization and mapping, or SLAM, problem [10],[15], [20]. In SLAM, the robot acquires a map of its environment while simultaneously estimating its own position relative to this map.

一般来说，由于地下全球位置信息的缺乏，使得矿井的测绘问题具有挑战性。因此，必须以同步定位和映射
的方式来处理地雷映射，或SLAM问题，[10]，[15]，[20]。在SLAM中，机器人获得环境的地图，同时估计
自己相对于地图的位置。

The SLAM problem is known to be particularly difficult when the environment possesses cyclic structure [5], [6], [13], [21]. This is because cycles pose hard correspondence problems that arise due to the (relatively) large position error accrued by a vehicle when closing cycles. Mines often contain a large number of cycles,hence the ability to handle cycles is essential for successful approaches to mapping mines.

This paper describes a SLAM algorithm for acquiring 3-D models of underground mines that can accommodate multiple cycles. Our algorithm uses a scan matching algorithms for constructing 2-D mine maps described in [14].
To close cycles, however, it utilizes an iterative correspondence algorithm based on the iterative closest point algorithm (ICP) [4], adapted to the problem of establishing correspondence in cyclic maps. 3-D maps are generated by applying scan matching to 3-D measurements after the 2-D mapping is complete.

Our algorithm has successfully enabled two robotic systems to acquire 3-D maps of mines. The first such system consists of an instrumented cart, which is pushed manually through a mine. This system is a low-cost solution to the mine mapping problem, but it can only be brought to bear in environments accessible to people. Our second system consists of a rugged robotic platform equipped with laser range sensors.

Abandoned mines, when dry, are often subject to low oxygen levels, poisonous gases, and they may be structurally unstable. Since bringing humans into such mines exposes them to a serious danger of life, the employment of autonomous robotic systems appears to be natural solution.

This paper provides results obtained in two different mines,both located in Pennsylvania, USA. One of these mines is a research mine, accessible to people. Another is a former
deep mine turned into a strip mine, inaccessible to people but accessible to robotic vehicles.