Toward automatic classification of chemical sensor data from autonomous underwater vehicles

Abstract

Autonomous underwater vehicles (AUVs) are commonly used to support oceanographic science by providing water-column mapping, seafloor bathymetric and photographic survey, and deep-sea exploration capabilities. In practice, the mapping activities carried out by AUVs consist of flying either pre-programmed tracklines (most propeller-driven AUVs), or else reporting data to human operators at regular intervals that permit retasking (typical for month-long underwater glider deployments). AUVs equipped with the ability to reason about scientific objectives in real time could significantly increase the value of individual deployments by enabling sampling efforts to be focused on targets or areas identified autonomously or semi-autonomously as scientifically interesting [1]. In this paper, we focus on AUV autonomy as it pertains to watercolumn sensing and argue that the classification of water-column sensor data represents an important enabling capability. We demonstrate practical, semi-supervised classification of watercolumn sensor data using a particular Bayesian, non-parametric clustering method, the Variational Dirichlet Process, combined with operator-supplied semantic labeling. The method is applied to the detection of a deep subsea hydrocarbon plume using data collected by the Woods Hole Oceanographic's Sentry AUV during an expedition to the Gulf of Mexico following the Deepwater Horizon blowout disaster.

Assessing the deepwater horizon oil spill with the sentry autonomous underwater vehicle

“…Furthermore, water velocity measurements obtained with the DVL on Sentry, showed that the water currents at 1100m matched the direction of the plume measured with the mass spectrometer and provided information on the total hydrocarbon flux contained within the plume. The scientific findings of this expedition are reported in [5] and [15] reports additional results on the classification of chemical sensor data obtained on these dives.…”

“…Two additional expeditions are planned for this year -the first to investigate naturally occurring methane seeps off of California and the second will use magnetic measurements to estimate the age of the Western Pacific tectonic plate. In addition, we are continuing to advance both the sensor payload and autonomy of the vehicle [15]. These efforts are not just focused on oceanography but, through an award from NASA, also on using Sentry missions as analogues for astrobiological exploration [6], [16].…”

Toward autonomous underwater mapping in partially structured 3D environments

“…Other environmental sensors mounted that have been used on AUVs include sheet lasers [75], fluorometers [15], mass spectrometers [36], optical backscatter sensors [100], and magnetometers [88]. If a prior map of su cient resolution is available, these environmental sensors can be used to navigate within the map.…”

Underwater 3D structures as semantic landmarks in SONAR mapping

“…While prior 3D mapping of the site provides a solution to the navigation problem by creating occupancy grids, enabling online obstacle avoidance and path planning, complex operations often require an analysis of the 3D map by a human operator to identify the elements of interest and the potential hazards [9]. This intervention considerably increases the operation time and stress on the human operator.…”