TY - JOUR
T1 - Active planning for underwater inspection and the benefit of adaptivity
AU - Hollinger, Geoffrey A.
AU - Englot, Brendan
AU - Hover, Franz S.
AU - Mitra, Urbashi
AU - Sukhatme, Gaurav S.
PY - 2013/1
Y1 - 2013/1
N2 - 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 an 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, non-adaptive 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.
AB - 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 an 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, non-adaptive 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.
KW - active perception
KW - adaptivity gaps
KW - motion planning
KW - sensor coverage
KW - underwater robotics
UR - http://www.scopus.com/inward/record.url?scp=84872738346&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84872738346&partnerID=8YFLogxK
U2 - 10.1177/0278364912467485
DO - 10.1177/0278364912467485
M3 - Article
AN - SCOPUS:84872738346
SN - 0278-3649
VL - 32
SP - 3
EP - 18
JO - International Journal of Robotics Research
JF - International Journal of Robotics Research
IS - 1
ER -