PipeSLAM: Simultaneous localisation and mapping in feature sparse water pipes using the Rao-Blackwellised particle filter

Abstract

Abstract-Water, a valuable resource, is usually distributed through urban environments by buried pipes. These pipes are difficult to access for inspection, maintenance and repair. This makes in-pipe robots an appealing technology for inspecting water pipes and localising damage prior to repair from above ground. Accurate localisation of damage is of critical importance because of the costs associated with excavating roads, disrupting traffic and disrupting the water supply. The problem is that pipes tend to be relatively featureless making robot localisation a challenging problem. In this paper we propose a novel simultaneous localisation and mapping (SLAM) algorithm for metal water pipes. The approach we take is to excite pipe vibration with a hydrophone (sound induced vibration), which leads to a map of pipe vibration amplitude over space. We then develop a SLAM algorithm that makes use of this new type of map, where the estimation method is based on the RaoBlackwellised particle filter (RBPF), termed PipeSLAM. The approach is also suited to SLAM in plastic water pipes using a similar type of map derived from ultrasonic sensing. We successfully demonstrate the feasibility of the approach using a combination of experimental and simulation data.

Robot Localization in Water Pipes Using Acoustic Signals and Pose Graph Optimization

“…The data in this work are similar in nature to that reported in our previous work [ 25 , 26 ]; however, it has been extended from a 0.4-m pipe to a 5-m pipe, shown in Figure 2 a. The process of acquiring the spatially varying signal is the same as in a previous work [ 25 ].…”

“…This quantitative result is difficult to compare to results found in the literature. The most comparable results are from previous work using an acoustic signal to build a map for localization in a pipe [ 26 ] and other work using a magnetic field to improve localization in a drilling robot [ 14 ]. In both cases, as in this work, the robot travels twice alone the same path and uses a continuous spatially varying field to improve localization.…”

“…An acoustic signal for robot localisation in water pipes was used in our previous work, using an Extended Kalman Filter [ 25 ] and a Rao-Blackwellised Particle Filter [ 26 ]. However, those studies used a short length of pipe, 0.5 m, whilst this study used a 5-m length of pipe.…”

“…In previous work, we have used a hydrophone emitter-receiver positioned on a robot to perform localization [ 25 , 26 ]. The hydrophone emits a sound wave that interacts with a metal pipe in a way that varies over space and so can be used to recognise location.…”

“…In order to evaluate the pose-graph optimization algorithm, we use data recorded in a 5-m long metal pipe filled with water, which is an order of magnitude longer than in our previous work [ 25 , 26 ]. We also analyse the performance of the algorithm in simulations over much larger scales than we could do in a laboratory, which relates more closely to real-world pipe networks.…”