A solution to the multiple aspect coverage problem

Bays, Matthew J.; Shende, Apoorva; Stilwell, Daniel J.; Redfield, Signe

doi:10.1109/icra.2011.5980222

Cited by 9 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The IMVP approach developed in this article is tested on a variety of target fields and compared to the state-of-the-art multiview planning methods known as MAC and clustered MAC (CMAC) [1], [2], [38], [69], [70]. Because the objects' locations and features used for classification all influence the UUV-based sensor performance, the IMVP approach is demonstrated first by considering different object layouts (Section VII-A) and, then, different classification sets (Section VII-B) using the simulation environment described in Section VI.…”

Section: Imvp Performance Resultsmentioning

confidence: 99%

“…In the following sections, the IMVP performance is demonstrated for a variety of target fields characterized by different layouts (Section VII-A) and classification features (Section VII-B). In every case study, the IMVP performance is compared to the MAC algorithm, which plans the shortest multiview path to cover every object using a fixed preplanned number of aspect angles, such that every object is detected at least once from each aspect angle [1], [38], [69], [70]. The MAC path may be inefficient for sparse object layouts, requiring the UUV-based sensor to travel long times without observing any objects [1], [38], [69], [70].…”

Section: Imvp Performance Resultsmentioning

confidence: 99%

“…In every case study, the IMVP performance is compared to the MAC algorithm, which plans the shortest multiview path to cover every object using a fixed preplanned number of aspect angles, such that every object is detected at least once from each aspect angle [1], [38], [69], [70]. The MAC path may be inefficient for sparse object layouts, requiring the UUV-based sensor to travel long times without observing any objects [1], [38], [69], [70]. The modification proposed in [2], known as CMAC, overcomes this limitation by designing the path based on the size of object clusters that may occur in applications with man-made TOIs [71], [72].…”

Section: Imvp Performance Resultsmentioning

confidence: 99%

“…Before a directional sonar, such as a forward-looking or sidescan sonar, is implemented, the ROI is typically surveyed using long-range sensors that provide a rudimentary estimate of target position and orientation. Subsequently, a directional sensor installed on a UUV is implemented to obtain multiple images or "looks" of every object, until a desired confidence level (CL) is achieved by the target classification algorithm [1], [38]. As a result, the time and power required to properly classify multiple targets are highly dependent on the UUV path.…”

Section: Problem Formulationmentioning

confidence: 99%

“…Previous multiview planning studies showed that path performance depends strongly on the object layout [1], [2], [38], [69], [70]. In particular, algorithms that perform well for objects uniformly distributed spatially, at random, over the ROI may not perform adequately when objects are laid out into clusters, and viceversa.…”

Section: A Influence Of Object Location On Imvp Performancementioning

confidence: 99%

See 4 more Smart Citations

Informative Multiview Planning for Underwater Sensors

Shin

Chang

Weaver

et al. 2022

IEEE J. Oceanic Eng.

View full text Add to dashboard Cite

Many modern imaging sensors must obtain multiple looks or "views" of a target at different orientations to automatically classify it with high confidence. Therefore, when tasked with classifying many targets, a mobile sensor may need to travel a long distance to change its position and orientation relative to every target, resulting in costly and time-consuming operations. This article presents a novel and general approach, referred to as informative multiview planning (IMVP) that simultaneously determines the most informative sequence of views and the shortest path between them. The approach is demonstrated on an underwater multitarget classification problem in which a sidescan sonar installed on an unmanned underwater vehicle must classify all targets in minimum time. Simulation and experimental results show that IMVP can achieve the same, or better, classification performance in half the time of existing multiview path planning methods. Also, by determining the most informative views and the shortest path between them, IMVP significantly improves classification efficiency, classification confidence level, as well as performance robustness.

show abstract