Uncertainty-driven view planning for underwater inspection
Open Access
- 1 May 2012
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 28 (10504729), 4884-4891
- https://doi.org/10.1109/icra.2012.6224726
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 sonar-derived 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.Keywords
This publication has 18 references indexed in Scilit:
- Consolidation of unorganized point clouds for surface reconstructionACM Transactions on Graphics, 2009
- Gaussian process modeling of large‐scale terrainJournal of Field Robotics, 2009
- Algorithms for subset selection in linear regressionPublished by Association for Computing Machinery (ACM) ,2008
- A Focus on Recent Developments and Trends in Underwater ImagingMarine Technology Society Journal, 2008
- A Vehicle System for Autonomous Relative Survey of In-Water ShipsMarine Technology Society Journal, 2007
- Vision Sensor Planning for 3–D Model AcquisitionIEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2005
- Active recognition through next view planning: a surveyPattern Recognition, 2003
- Chained Lin-Kernighan for Large Traveling Salesman ProblemsINFORMS Journal on Computing, 2003
- Information theoretic sensor data selection for active object recognition and state estimationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2002
- Active visionInternational Journal of Computer Vision, 1988