A Vehicle System for Autonomous Relative Survey of In-Water Ships

2012 IEEE International Conference on Robotics and Automation

et al. 2012

Abstract-We discuss the problem of inspecting an underwater structure, such as a submerged ship hull, with an autonomous underwater vehicle (AUV). In such scenarios, the goal is to construct an accurate 3D model of the structure and to detect any anomalies (e.g., foreign objects or deformations). We propose a method for constructing 3D meshes from sonarderived point clouds that provides watertight surfaces, and we introduce uncertainty modeling through non-parametric Bayesian regression. Uncertainty modeling provides novel cost functions for planning the path of the AUV to minimize a metric of inspection performance. We draw connections between the resulting cost functions and submodular optimization, which provides insight into the formal properties of active perception problems. In addition, we present experimental trials that utilize profiling sonar data from ship hull inspection.

Section: Building 3d Modelsmentioning

confidence: 99%

Uncertainty-driven view planning for underwater inspection

Hollinger

Englot

2012 IEEE International Conference on Robotics and Automation

et al. 2012

“…A Kalman filter is used to produce depth and attitude estimates using the IMU and depth sensor, and DVL velocities are integrated to provide x-y position estimates. More details on the localization system are provided in Hover et al (2007).…”

Section: Surface Reconstruction and Inspection Planningmentioning

confidence: 99%

“…In this domain, an AUV is inspecting the submerged portion of a ship hull, and the goal is to plan the actions of the AUV to gather data that results in the best 3D reconstruction of the hull. We utilize the Bluefin-MIT Hovering Autonomous Underwater Vehicle (HAUV), which was designed for autonomous surveillance and inspection of in-water ships (Hover et al, 2007). The proposed algorithm follows the following steps, which are discussed in more detail in the subsequent sections:…”

Section: Surface Reconstruction and Inspection Planningmentioning

confidence: 99%

Active planning for underwater inspection and the benefit of adaptivity

Hollinger

Englot

The International Journal of Robotics Research

et al. 2012

133

We discuss the problem of inspecting an underwater structure, such as a submerged ship hull, with an autonomous underwater vehicle (AUV). Unlike a large body of prior work, we focus on planning the views of the AUV to improve the quality of the inspection, rather than maximizing the accuracy of a given data stream. We formulate the inspection planning problem as an extension to Bayesian active learning, and we show connections to recent theoretical guarantees in this area. We rigorously analyze the benefit of adaptive re-planning for such problems, and we prove that the potential benefit of adaptivity can be reduced from exponential to a constant factor by changing the problem from cost minimization with a constraint on information gain to variance reduction with a constraint on cost. Such analysis allows the use of robust, nonadaptive planning algorithms that perform competitively with adaptive algorithms. Based on our analysis, we propose a method for constructing 3D meshes from sonar-derived point clouds, and we introduce uncertainty modeling through non-parametric Bayesian regression. Finally, we demonstrate the benefit of active inspection planning using sonar data from ship hull inspections with the Bluefin-MIT Hovering AUV.

“…In our experiments we use the Hovering Autonomous Underwater Vehicle (HAUV) [20] from MIT/Bluefin (figure 6). It is equipped with five rim-driven thrusters that make it directly controllable in all axes but roll.…”

Section: A Platformmentioning

confidence: 99%

Underwater inspection using sonar-based volumetric submaps

Teixeira

Kaess

2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

et al. 2016

Abstract-We propose a submap-based technique for mapping of underwater structures with complex geometries. Our approach relies on the use of probabilistic volumetric techniques to create submaps from multibeam sonar scans, as these offer increased outlier robustness. Special attention is paid to the problem of denoising/enhancing sonar data. Pairwise submap alignment constraints are used in a factor graph framework to correct for navigation drift and improve map accuracy. We provide experimental results obtained from the inspection of the running gear and bulbous bow of a 600-foot, Wright-class supply ship.