Home | Contact ST  
Follow ST



March 2014 Issue

TRIDENT: Marine Robots for Autonomous Intervention Missions
Pedro J. Sanz

TRIDENT is a research project launched in 2010 and funded by the European Commission to increase robot autonomy levels for underwater intervention missions. The project brings together marine research skills in navigation and mapping for underwater robotics, multisensory perception, and techniques related to intelligent control architectures, vehicle manipulator systems and dexterous manipulation.

During the last 20 years, AUVs have become a standard tool for mapping the seafloor using optical and acoustic sensor modalities, with applications to dam inspection, marine geology and underwater archaeology to mention but a few. Most available autonomous platforms can perform side scan sonar and bathymetric multibeam surveys.

There are a large number of potential AUV applications beyond surveying, e.g., the maintenance of permanent observatories, submerged oil wells, cabled sensor networks, pipes, the deployment and recovery of benthic stations, and the search and recovery of black boxes. Nowadays, these tasks require the use of work-class ROVs deployed from dynamic positioning vessels, making them very expensive. To handle these new applications, research to increase the autonomy of underwater intervention systems started in the early '90s with the pioneering AUVs Otter, Odin, Union and Amadeus, but field demonstrations were not conducted until the first decade of the 21st century.

The first fully autonomous intervention at sea was demonstrated by the ALIVE project when a hovering-capable AUV was able to home into a subsea intervention panel using imaging sonar and then dock into it with hydraulic grasps using visual feedback. Once attached to the panel, simple fixed-base manipulation was used to open/close a valve.

The first object manipulation from a floating vehicle (I-AUV) was achieved in 2009 by the SAUVIM project, which demonstrated the capability of searching for an object whose position was roughly known a priori. The object had artificial landmarks, and the robot autonomously located and hooked it with a recovery device while hovering.

The first AUV multipurpose object search and recovery strategy was demonstrated by TRIDENT in 2012. First, the object was found in a water tank using a down-looking camera and photo-mosaicking. Next, the object was autonomously 'hooked.' The experiment was repeated in a harbor environment using a 4-degrees-of-freedom arm, and later a 7 degrees-of-freedom arm with a three-fingered hand.

TRIDENT proposed a methodology for multipurpose underwater intervention tasks where a team of two cooperative heterogeneous robots with complementary skills, an autonomous surface craft (ASC) and an intervention autonomous underwater vehicle (I-AUV) with a dexterous manipulator, are used for manipulation tasks.

The methodology has two stages. In the first stage, the I-AUV is deployed from the ASC to execute a pre-plotted survey, gathering optical/acoustic data from the seafloor while the ASC provides georeferenced navigation data and communication with the end user. During this stage, the I-AUV carries out accurate path following and terrain tracking to maximize bottom coverage and data quality. The motion of the ASC is coordinated with that of the I-AUV to achieve precise ultrashort baseline positioning and reliable acoustic communications. After the survey, the I-AUV docks with the ASC and sends the data to a ground station, where a map is created and a target object is identified by the end user.

In the second stage, the ASC navigates to a waypoint near the intervention area where the I-AUV is launched to search for the object. When the target object has been found, the I-AUV switches to free-floating navigation mode. A dexterous hand attached to a redundant robot arm manipulates the object, assisted by perception technology. Particular emphasis has been made on the research of vehicle intelligent control architecture to provide the embedded knowledge representation framework and high-level reasoning required for autonomy and onboard decision making of the robot platform.

The new methodology allows the operator to specify an intervention task among a set of predefined ones to be performed on a specific target selected by the end user from a previously created map. Hence, the intervention task is a semi-automatic process: A human selects the target, which is automatically recognized and autonomously manipulated by the robot.

TRIDENT has demonstrated a new benchmark for search and recovery of submerged black boxes in shallow waters up to 10 meters with two sea trials. The first one was held in October 2011 at Roses Harbor (Girona, Spain) and the second in October 2012 in Port de Sóller naval station (Mallorca, Spain). Before the final TRIDENT validation experiments at Sóller, trials for integration and testing control were conducted at the University of Girona in September 2012. The complete mechatronics were successfully tested, and all the control components were validated. In October 2012, the full system was integrated and fully tested at sea. The target black box was located and recovered, and the program's objectives were accomplished.

Companies working in underwater scenarios with tasks requiring intervention capabilities could be the main beneficiaries of the know-how developed by TRIDENT. Whichever underwater task requiring manipulation, like valve opening, button pushing, cable connecting, seabed sampling or object recovery, can now be solved in a more efficient, secure, robust and cheap way. Potential end users are offshore companies, maritime rescue organizations, and coastal and ocean observatories.

Pedro J. Sanz is a full professor at Jaume I University in Spain, and head of the Interaction and Robotic Systems Lab. His research interests are multisensory-based grasping and dexterous manipulation, telerobotics and human-robot interaction applied to real-world scenarios, including assistive and underwater robotics. He was the TRIDENT project coordinator from 2010 to 2013.


-back to top-

Sea Technology is read worldwide in more than 110 countries by management, engineers, scientists and technical personnel working in industry, government and educational research institutions. Readers are involved with oceanographic research, fisheries management, offshore oil and gas exploration and production, undersea defense including antisubmarine warfare, ocean mining and commercial diving.