Abstract A new tool for subsea inspection in the offshore oil and gas industry iscurrently going through performance and qualification testing. The AutonomousInspection Vehicle (AIV) has been designed and built by Subsea7 and SeeByte Ltdto provide the industry with a valuable tool capable of making a positivecontribution to Life of Field operations. The station keeping and hovering ability of the AIV is the next step in theevolution of autonomous systems in the marine environment. Survey classautonomous vehicles have already shown their value with improved data qualityand efficiencies over traditional methods. The first commercial AIV will becapable of many of the inspection tasks currently carried out by RemotelyOperated Vehicles (ROV). Regular inspection data of risers, pipelines andseabed equipment can be gathered using a single AIV operating directly from anoffshore facility. A more rapid assessment of a field can be made usingmultiple systems operating together from a single support vessel. This has notbeen done before with a commercial vehicle and hence is truly leading edgetechnology. The paper will outline some of the technical challenges in creating the vehicleand how the use of advanced simulation linked to practical testing is beingused to ensure the performance of the system. Finally the paper will draw aparallel with the evolution of subsea infrastructure that has fully enabled thecapability of the ROV and demonstrate how the introduction of autonomoustechnology should be considered with confidence. Introduction Subsea 7 is a major engineering and construction company supplying a range ofservices and technologies primarily into the offshore oil and gas industry. Thecompany operates a worldwide fleet of high specification subsea constructionand Inspection vessels together with over 100 ROVs. The company is activelyinvolved in bringing new and innovative technologies to the market to supportits operational focus. The development of autonomous underwater vehicles is onesuch technology. SeeByte is a cutting-edge software company that develops some of the world'smost advanced software for managing unmanned and remote assets. The company isfocused on advancing autonomous vehicle development and is actively involved inmajor programs of work for international customers. SeeByte brings a proventrack record of de-risking and demonstrating this technology for customers inmilitary, offshore and subsea sectors.

This research addresses the problem of coordinating multiple autonomous underwater vehicle (AUV) operations. An intelligent mission executive has been created that uses multiagent technology to control and coordinate multiple AUVs in communication-deficient environments. By incorporating real-time vehicle prediction, blackboard-based hierarchical mission plans, mission optimization, and a distributed multiagent–based paradigm in conjunction with a simple broadcast communication system, this research aims to handle the limitations inherent in underwater operations, namely poor communication, and intelligently control multiple vehicles. In this research, efficiency is evaluated and then compared to the current state of the art in multiple AUV control. The research is then validated in real AUV coordination trials. Results will show that compared to the state of the art, the control system developed and implemented in this research coordinates multiple vehicles more efficiently and is able to function in a range of poor communication environments. These findings are supported by in-water validation trials with heterogeneous AUVs. © 2010 Wiley Periodicals, Inc.

Autonomo to gain routine and permanent access to the underwater environment. However, their uses remain constrained to very specific tasks with little real autonomy. Mission plans are generally static and scripts based and adaptation to the environment or components failure is very limited. This is particularly problematic in the case of multiple vehicles where knowledge discovery is rarely shared, responses to sensor or platform failure rarely dealt with and architectures for collaborative planning in the presence of low communications bandwidth is a real challenge. Finally, as these platforms get more widely used by increasingly operational military personnel, there is a real need to move from script based planning to goal based planning based on a clear description of each platform capabilities and requirements which can be used by an automated or human planner to achieve complex missions. The challenges related to this domain require the development and integration of more evolved embedded tools that can raise the platform's autonomy levels while maintaining the trust of the operator. This paper shows how knowledge representation including world models, reasoning, planning tools and mission executive can provide the required interoperability between embedded agents to achieve high-level mission goals described by the operator.

This research addresses the problem of coordinating multiple autonomous underwater vehicle (AUV) operations. A mission executive has been created that uses multi-agent technology to control and coordinate multiple AUVs in communication deficient environments. By incorporating a simple broadcast communication system in conjunction with real time vehicle prediction this architecture can handle the limitations inherent in underwater operations and intelligently control multiple vehicles. In this research efficiency is evaluated and then compared to the current state of the art in multiple AUV control.

The current state of the art in autonomous underwater vehicle (AUV) mission representation is sequential script-based plans. Though simple and easy to understand by an experienced user, this way of representing a mission is limited to static goal order and cannot easily handle unforeseen events. This limitation is amplified in multi-vehicle missions where actions need to be coordinated. This research aims to improve upon the current state of the art by designing and implementing a dynamic, hierarchical mission representation system based around the principles of blackboard systems and specifically designed to facilitate multi-vehicle coordination. The functionality of the system is tested in a simulated multiAUV mine countermeasures mission and efficiency is compared to the state of the art. Simulated results are then validated in real world trials with two AUVs.

An AUV has been designed and built in the Ocean Systems Laboratory at Heriot-Watt University to compete in the 2007 SAUC-E competition. Using a robust hardware and software design the vehicle is able to successfully complete the tasks set out in the competition and will be an excellent platform for further development.

This research addresses the problem of coordinating multiple autonomous underwater vehicle (AUV) operations. An architecture has been created that uses multi-agent technology to control and coordinate multiple AUVs in communication deficient environments. By incorporating a simple broadcast communication system in conjunction with real time vehicle prediction this architecture can handle the limitations inherent in underwater operations and intelligently control multiple vehicles. In this research efficiency in terms of mission speed, battery life, communication robustness and goal redundancy is evaluated and then compared to the current state of the art in multiple AUV control.

The force-multiplier advantages of multi-vehicle operations are well known, but to achieve the full benefits, suitable architectures that facilitate true cooperative autonomous planning must be realised. This study aims to review potential approaches from the existing literature on cooperative robotics, and through prototyping on real UxVs, de-risk the transition and future development of suitable technologies.

The force-multiplier advantages of multi-vehicle operations are well known, but to achieve the full benefits, suitable architectures that facilitate true cooperative autonomous planning must be realised. This study aims to review potential approaches from the existing literature on cooperative robotics, and through prototyping on real UxVs, de-risk the transition and future development of suitable technologies.

GENCEM is a genetic algorithms approach to coordinated exploration and mapping with multiple autonomous robots. Building on previous work in coordinated mapping, the work reported here compares static to evolutionary approaches for the same coordination tasks. In GENCEM, parameters affecting the coordination behaviors are evolved, leading to a decided improvement over hand-coded parameter settings across a variety of environments and using different numbers of robots. The success of this preliminary study demonstrates the viability of this approach for learning to coordinate, representing the first stage of implementation of a larger system for more complex coordination tasks and strategies.

The goal of this project was to develop a genetic algorithms approach for robotic agents to learn to communicate and coordinate with each other, so that the group can learn to be more effective at mapping unfamiliar terrain and locating items within that terrain.