Showing papers on "iRobot Seaglider published in 2000"

Oceanographic Surveys with a 50 hour Endurance Autonomous Underwater Vehicle

Abstract: Autonomous Underwater Vehicles (AUVs) are becoming accepted data-gathering tools within the marine science community in Europe, the US and elsewhere. Technology can now provide vehicles with a useful range and depth envelope. For example, the Southampton Oceanography Centre’s Autosub-1 vehicle has already covered 263 km in a single oceanographic survey mission, reaching depths of 500 m off Bermuda in September 1998. The challenge now is to ‘free the technology from the research community. Potential direct and indirect benefits from the offshore energy industry’s use of AUVs have been quantified. For one company, the benefits of using survey class AUVs has been estimated at $60m over 5 years. The economics become more attractive as the industry moves to deeper water. While today’s AUVs of long endurance are limited in depth, developments on the horizon in composite materials and high capacity, lightweight secondary batteries will enable 1000 km range, eight-day endurance and at least 1600 m diving depth to be in active service within the next two years. This paper will review the oceanographic survey achievements of the Autosub AUV over its 216 missions (to December 1999), working in UK, US and Bermudan waters. Moving to the future, the paper will illustrate how such an AUV could be used in the offshore industry for survey, monitoring and emergency response. Finally, we will describe the present Autosub development programme, scheduled to deliver an operational 1000 km, 1600 m vehicle by mid 2000. Introduction The Autosub AUV was conceived as a survey vehicle to complement research ship programs by the provision of a cost effective tool to collect water column and seafloor information. Included within the design brief was the aim to address the widest possible scientific community hence allowing varied missions and payloads. This requirement ensured that flexibility was fundamental to the technology design. Hence for Autosub-1 there was no fixed sensor suite, the vehicle sub-systems were modular and thus offered a simple upgrade path as new technologies became available. One other by-product which is envisaged but not yet proven is the opportunity to change the vehicle characteristics i.e. size and shape, without a major re-design program. (Fig 1) The first vehicle, Autosub -1 was constructed in 8 months, being completed in May 1996. The short production cycle was only possible because of the extensive development programme which had been underway for the previous 5 years. The first phase saw Autosub-1 progress from its first ‘in water’ missions in Empress Dock, Southampton, (June 1996) to completion of the acceptance trials in just under one year (April 1997) Oban 97 The Autosub performance and acceptance trial criteria have been described by Millard et al (1), the main objectives were: • Profiling autonomously from close to the surface to near the seabed in at least 100m of water; • Mission lengths of at least 6 hours, with satellite fixes updating the dead reckoning(DR) vehicle navigation; and • Collection of meaningful scientific data. After completion of the engineering acceptance phase, the vehicle program has been strongly focussed on the customers’ scientific objectives, which has taken the vehicle into a multirole capability, via a series of campaigns worldwide. Florida 97 Florida in December 1997 was a significant step, as it moved Autosub from missions in the sheltered coastal waters off Oban, Scotland to the open seas of the Atlantic Ocean. In a joint collaboration with Florida Atlantic University (FAU), co-funded by the US Office of Naval Research (ONR), a campaign saw the vehicle deployed on the edge of the Gulf Stream. Local conditions, with surface current speeds of up to 2m/s, provided significant challenges for the vehicle operating a water column regime of above average current shear and sound speed gradients. These environmental features also OTC 12003 Oceanographic surveys with a 50 hour endurance autonomous underwater vehicle G Griffiths, K G Birch & The Autosub Technical Team :N.W. Millard, S.D. McPhail, P. Stevenson, M. Pebody, J. R. Perrett, A.T. Webb, M Squires & A Harris, Southampton Oceanography Centre 2 G Griffiths, K G Birch & the Autosub Technical Team OTC 12003 provided the support team with a challenge to keep track of the vehicle’s progress throughout individual missions. The results of the campaign have been described by Griffiths et al (2), the main achievements were: • 72 hour mobilisation from unpacking the container to operational deployment of the vehicle into the water; • completion of a 110km CTD & ADCP survey; • terrain following mission 10m above seabed in water depths varying between 12-204m; and • due to high surface current the vehicle was out of acoustic contact with the support boat – thus demonstrating unsupervised operations. Bermuda 98 With deployment and recovery via the Bermuda Biological Station vessel, the Weatherbird, Autosub was a key component in the Autonomous Vehicle Validation Experiment (AVVEX). The overall aim of the experiment was to demonstrate the synergy between a moored times series dataset and an AUV spatial survey. Amongst the challenges for Autosub were to: • increase the range and depth capability; • demonstrate the easy of integration for commercial off the shelf (COTS) sensors; • gather multidisciplinary data in a number of survey modes; and • verify that operations could become routine and cost effective. The campaign had proved once again the viability of transporting the vehicle, plus the support team, and the ability to work on different vessels without significant time penalties to mobilise. During AVVEX the vehicle completed the longest mission to date, covering 263km and routinely profiling to a depth of 400m. The mission lasted approximately for 53 hours, during which the recovery waypoint was altered, by radio link, to provide safer operational conditions for the Weatherbird recovery as the sea state had worsened during the mission time.(Sea Technology Feb 2000) The vehicle also achieved a record descent to 504m. As a result of the 263km mission it was possible for the first time to establish a the vehicle battery cost as 43$ per MJ, when using primary manganese alkaline batteries. This equates to about 15$ per kilometre, and compares well with silver zinc re-chargeable cells at 30$ per MJ, assuming a maximum life of 20 cycles. The comparison excludes the staff time to manage the cell re-charging. One area this did highlight was the need for a dedicated launch and recovery system. The height of the Weatherbird ‘A’ frame with the associated long lifting strop, meant that launching was difficult, but recovery was far from being a routine operation. Vehicle flexibility as shown on Scotia & Challenger campaigns A critical feature for Autosub to meet the variation in research programs is the ability to change vehicle data gathering suite between campaigns. This of course must be achieved in a timely and cost effective manner. However by changing the sensor payload there are inevitably other consequential effects on the overall vehicle, for example a change in buoyancy distribution. With this in mind a key strength within the Autosub design is the systems modularity, with individual sub-systems interconnected to the mission controller, via the Lonworks serial network. The modular approach also simplifies the addition, or removal, of sensors etc, and further reduces the risk by allowing each sub-system to be tested and proven in a ‘stand alone mode’ before integration into the vehicle. This also helps to minimise the vehicle downtime, when changing between mission requirements. Proof of this flexibility was highlighted when the vehicle took part in two campaigns in quick succession in August 1999, which necessitated a change of sensor suite, and also of survey vessels. The first of a pair of programs was the ’Under Sea Ice and Pelagic Surveys’ (USIP) undertaken for Dr Paul Fernandes of the Marine Laboratory, Aberdeen, and Drs Andrew Brierley and Mark Brandon of the British Antarctic Survey, onboard the RV Scotia. The scientific objective being to understand differences between datasets collected by a research vessel and AUV, both using exactly the same EK500 fisheries echo sounder. The inter-comparison programme involved the vehicle working in several different modes to test various hypotheses. For 5 missions the vehicle worked directly within acoustic range of the Scotia, whereas for 8 missions Autosub was ‘unescorted’, and out of communication range with the vessel. We believe this is the first time that any AUV has undertaken planned ‘unescorted’ missions away from a support vessel. Griffiths et al (3). Another success during this campaign was the first operational use a custom designed ‘launch and recovery’ gantry. This enabled the vehicle operations over the starboard side of the Scotia, overcoming the problems encountered during the AVVEX campaign. Gantry operations also removed the requirement for small boat deployment, increasing the weather window for deployment and recovery. The challenge after completing the Scotia programme was to move the operations to the RRS Challenger to undertake turbulence measurements over Linear Sandbanks for George Voulgaris, of the University of South Carolina. The switch between vessels, demobilising, changing the primary sensor suite, and mobilisation was all achieved within a space of three weeks, this was no mean task. The campaign saw the vehicle, operating in bottom terrain following mode, and again undertaking some of the missions in an ‘unescorted’ mode. Programme for 2000 and onwards This year will see further campaigns within the NERC Science Mission program with: OTC 12003 Oceanographic surveys with a 50 hour endurance autonomous underwater vehicle 3 • Sonar and turbulence studies of the west coast of Scotland; • Measurement of dissolved particulate manganese in two Scottish sea lochs;

An Underwater Vehicle Monitoring System and Its Sensors

Abstract: This paper describes a virtual collaborative world simulator, DVECS (Distributed Virtual Environment Collaborative Simulator), for underwater robots and its underwater vehicle, SAUVIM (Semi-Autonomous Underwater Vehicle for Intervention Missions). DVECS is used for testing unmanned underwater vehicles (UUVs) of both real and simulated worlds where interaction and cooperation of other real and simulated vehicles, obstacles, situations, conditions and disturbances in a hybrid, synthetic, virtual environment can be observed without physical intervention. This virtual system can be used to determine: (1) the optimal performance and criteria for the cooperating vehicles and its relative application; (2) the determination of the advantages and disadvantages of collaborative application tasks between multiple UUVs; and (3) the optimal communication links between the cooperating vehicles and its remote control stations. DVECS is used as a monitoring system.

Direction of Development of Autonomous Underwater Vehicles

Abstract: The autonomous underwater vehicle (AUV) has been recognized as a new platform for observing the underwater world. It can dive freely around the target of the mission prescribed in its computer prior to launching.Throughout a dive, it does not require the operator and the support vessel to help it to execute its mission. It has no umbilical cable, which is fitted to remotely operated vehicles currently used for underwater operations and observations and usually causes problems for the operator. This paper presents the state of the art of the AUV and how it can be used in underwater science.