How robots can reduce accidents in offshore energy asset inspection
05 October 2020
As industry and academia work together to improve safety in the offshore energy sector, the ORCA Hub is creating game-changing robotic solutions for asset inspection, maintenance and repair.
An autonomous drone from Imperial College London – Image: ORCA Hub
(Click here to view article in digital edition)
Working offshore presents daily challenges for those inspecting and maintaining assets in an unpredictable and often hazardous work environment. Changing weather patterns, falls from height, the dangers of working under the sea and on platforms combine to create significant dangers for the humans that deliver the industry’s stringent health and safety recommendations. The renewables sector also recognises the dangers its workers face as it strives to keep them safe offshore.
To address these safety concerns, a consortium of five UK universities and more than 30 industry partners has joined together. Using ground-breaking research to inform and create industry ready technology, the solutions are designed to support those working in the field and minimise the risk posed to human life.
Introducing ORCA Hub
The Offshore Robotics for Certification of Assets (ORCA) Hub is the largest academic centre in the world for research into robotics technology for offshore energy infrastructure. Its aim is to remove humans from hard to reach, hazardous and dangerous work environments by advancing Robotics and Artificial Intelligence (RAI) technologies for the inspection, repair, maintenance and certification of offshore energy platforms and assets. Its solutions are designed for use in both the oil and gas and renewable energy fields.
Launched in 2017 and led by the Edinburgh Centre for Robotics, a partnership between Heriot-Watt University and the University of Edinburgh, the consortium includes the University of Liverpool, Imperial College London and the University of Oxford. In 2020 an additional 8 UK-based institutes have joined through the Partnership Resource Fund. The fund enables the Hub to expand its current work into new areas with a clearly defined industrial need.
In close collaboration with industry, ORCA’s researchers are developing robotic technologies that are reliable, robust and certifiable. The innovations can inspect assets, make autonomous decisions and, in some cases, carry out repairs – overseen by human operators onshore. The focus of these activities is solving the challenges that are inherent in the extreme and unpredictable daily environment of the offshore energy workplace.
Verification and validation (V&V) – Image: ORCA Hub
With solutions deployable in the air, on the surface, and underwater, ORCA’s goal is to create remote solutions that can readily integrate with existing and future assets and sensors. Designed specifically to operate and interact safely in autonomous or semi-autonomous modes in complex and cluttered environments, ORCA’s potential contribution to improving standards of health and safety offshore is significant.
With the proposition being that robots will make decisions and carrying out actions autonomously, a new problem inevitably occurs. If a robot is sent out to inspect and repair an offshore asset and ends up damaging it further, or even worse hurts a human operator, this is an unacceptable risk. How can an autonomous robot be trusted to make the right decision and carry out actions safely?
A question of trust
Dr Vincent Page and Dr Matt Webster from the University of Liverpool explain how they are working towards building trust in a robot’s decisions and actions.
Significant question marks remain over the widespread use of Robotic and AI (RAI) solutions in the sector. A specific concern is whether or not industry and individual operators can be sure that robots will operate safely, particularly in dynamic and uncertain environments. This is particularly true for Beyond Visual Line of Sight (BVLOS) situations, where the human operator cannot see the robotic system in operation.
The offshore energy industry has both a human workforce as well as high-value assets to protect so, by working closely with regulators, the University of Liverpool team is creating methodologies that can help to build confidence in the operation of the proposed autonomous RAI systems.
The research efforts focus on robot certification as part of ORCA’s wider collaboration. To certify autonomous systems, Verification and validation (V&V) is used, which is a set of methodologies that ensures robots make correct and reliable decisions and carries out safe actions for a given task.
HDT manipulator – Image: ORCA Hub
Any autonomous robot that is designed and built for offshore environments should be certified before deployment. Regulations will eventually state that autonomous systems must meet specific requirements, including minimum safe distance tests and failure responses. However, these standards are yet to be put in place. More work needs to be done to provide a set of regulations to guide the private sector’s activity and investment. Most regulations assume that there will always be a human in the loop, where for autonomous systems, this is not always the case.
To help achieve this, the Liverpool team is creating a new approach termed Corroborative Verification & Validation. This allows a combination of different V&V techniques to be used together to generate evidence that a particular system is safe or reliable. The evidence from each V&V technique can be corroborated by the other V&V techniques, providing a higher level of confidence in the behaviour of autonomous robots prior to deployment in the field.
At present, corroborative V&V on ORCA Hub research employs a three-pronged approach, based on formal methods, simulation-based testing and real-world experiments. This allows the research team to put the robotic system through a large number of potential operational scenarios, including those that are potentially risky. This helps to create a well-rounded picture of the capabilities of the autonomous robotic system. By following this in-depth approach, it is possible to determine with higher confidence that the robots are making the right choices and that these choices result in safer actions.
The real-world experiments use an autonomous unmanned rotorcraft. Simulation-based testing is achieved using a distributed computer simulation of the aircraft engaged in an offshore inspection task. Finally, formal methods are used to exhaustively analyse the aircraft’s autonomous decision-making system. The use of Corroborative V&V allows the research team to use the best features of each V&V technique to strengthen the argument that an autonomous robot is safe to use: the intuitive and powerful impact of real-world experiments, the detailed nature of computer simulation, and the exhaustiveness of formal methods.
While essential, the certification process can be costly in terms of time and labour. However, by using Corroborative V&V, it is possible to increase the level of confidence in the results while reducing the time and effort required to certify an autonomous system for use in real-world deployments. The technique can also be applied in a similar way to other kinds of robotic systems and is not limited to the current case study.
The Liverpool team can answer many fundamental questions about the autonomous system: Does it make safe decisions? Does it carry out safe actions? Do these answers match reality? The team can then have greater confidence that an autonomous system is ready to be deployed safely and reliably.
ANYmal – Image: ORCA Hub
To build further confidence, the ORCA Hub regularly demonstrates the application of its robotic solutions to industry partners. These capability demonstrations have taken place in the waters of Loch Linnhe, on a mock oil rig at a Fire Training College and, most recently, at the Offshore Renewable Energy Catapult test facility in Blyth.
The showcases demonstrate the Hub’s four main focus areas and how they support industry: mapping, surveying and inspection; planning control and manipulation; human-robot interaction; and robot and asset self-certification.
Why is ORCA’s research needed?
As a new decade begins, the international offshore energy industry faces many challenges, including near-permanent low oil prices and expensive decommissioning commitments from historic infrastructure, particularly in the North Sea. The dangers that humans continue to face while inspecting and repairing energy infrastructure are significant, despite every effort to minimise the risks involved.
Professor David Lane, director of the ORCA Hub, said: “The ORCA Hub’s game-changing, remote solutions will help the UK to export agile products and services internationally while ensuring our energy sector remains both economically viable and globally competitive. Technology is becoming increasingly important in safeguarding humans from hazardous work environments and the UK is at the forefront of this revolution.”
For more information about ORCA Hub, visit www.orcahub.org
Contact Details and Archive...