Towards Persistent Real-Time Autonomous Surveillance and Mapping of Surface Hydrocarbons
Anni K. Vuorenkoski, Jesse C. Thomas, Fraser Dalgleish, Bing Ouyang, Peter D Carragher
- Year
- 2013
- Citations
- 2
Abstract
Abstract This paper describes results and findings from a recent two month robotic surface seep monitoring mission in the north-eastern Gulf of Mexico. The studied sites are at Mississippi Canyon salt domes and mud volcanoes which have well known natural seepage, resulting in the existence of regular but short-lived surface oil slicks. Effective real-time monitoring of such phenomenon requires new in situ technology that is capable of long term persistent surveillance and robust autonomous hydrocarbon detection. The core component of this mission was the Wave Glider autonomous marine vehicle integrated with a multi-channel optical sensor payload and weather sensing instrumentation package. A statistical process control (SPC) based automatic hydrocarbon event detection algorithm was also further developed and demonstrated during the mission. The paper also includes results describing the Wave Glider multi-sensor signatures to known hydrocarbon concentrations in typical marine elastic and inelastic scattering background environments. A series of experiments were independently conducted in both controlled laboratory conditions and also in a controlled spill facility, with ground truth spectrofluorometry and water sampling analytics results also being obtained. The event detection algorithmic approach was tested and refined using data from these sets of experiments, as well as the data acquired during the field mission. The algorithm was executed with near real-time Wave Glider data during the second month of the mission and results demonstrate that robust real-time surface hydrocarbon detection is possible and can be used to reconfigure the Wave Glider behavior, converting from a surveillance mode to a survey or mapping mode in order to determine the boundaries and origin of the identified surface slicks. To further corroborate with the in situ results, coincident SAR imagery collects were made at certain phases of the mission to assist with the analysis of the multi-sensor signatures, slick lifetime and spatial extent as derived by the Wave Glider measurements.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991