ORCA Hub Open Day
This week we hosted an ORCA Hub Open Day at Heriot-Watt University to share information on some of the Hub’s activities over the first year of the project and our plans going forward. The event drew attendees from the oil and gas and renewables industries, funding bodies, innovation centres, development agencies and other universities and was hosted at the Edinburgh Business School and ORIAM Scotland. Presentations from Hub Director, David Lane, and the Work Theme leads were followed up by an exhibition of demonstrations and posters from the ORCA teams in the afternoon.
The following demonstrations were on display:
1. The Limpet: a multi-sensing, immobile robot – Alistair McConnell, Sayed Mohammed (University of Edinburgh)
The Limpet is designed to monitor the local environment over an extended period of time in remote and hazardous locations. These locations include offshore and onshore energy platforms.
2. Four-legged robot walking in rough terrain – Wouter Wolfslag, Guiyang Xin, Carlo Tiseo, Michael Mistry (University of Edinburgh)
Offshore platforms can be hard for robots to navigate: they have cluttered areas, stairs and their walking surfaces can be slippery. Four-legged robots promise a way to handle these challenges.
3. Navigation and motion planning with the ANYmal quadruped – Andreia Vasconcelos, Simona Nobili, Milad Remanazi (University of Oxford)
New methods for planning paths and navigating the ANYmal in complex and realistic environments were shown and the team gave an overview of their upcoming field trails at the Fire Service College (Gloucestershire).
4. Underwater 3D reconstruction and mapping with stereo vision – Tomasz Luczynski, Sen Wang (Heriot-Watt University)
Cameras are powerful sensors that not only provide colour images but can also be used to reconstruct a 3D model of the observed scene.
5. Operator cognitive load estimation through language – Helen Hastie, José David Lopes, Katrin Lohan (Heriot-Watt University)
The cognitive load imposed by the task of managing multiple vehicles in stressful situations has been shown to affect the way people communicate, their actions and their information needs. In order to provide a multimodal interface (such as MIRIAM) that is in-tune with the user, we need to monitor and adapt to the user state.
The ORCA Hub is working on predicting cognitive overload simply using the operator’s language and as such in a truly non-invasive way, compared to previous methods such as EEG. We have studied ‘symptoms of cognitive load’ in terms of what a user says, the types of words they use and how they say it (intonation). The future goal on ORCA is to be able to track these symptoms during interactions and have the ORCA multimodal interface adapt to the user in real-time, for example, delaying less urgent information or re-prioritising tasks. This work moves towards more efficient collaboration between humans, robotics and autonomous systems on offshore energy platform maintenance tasks.
6. Video commentary summarisation, search and indexing using topic mapping – Mike Chantler, Pierre Le Bras (Heriot-Watt University)
This demonstration presented Topic Mapping; an interactive visualisation system that provides intuitive overviews of large collections of unstructured document collections and voice corpora at arbitrary levels of detail. It also has the potential to turn the narratives in reports (e.g. drilling well surveys etc.) into actionable feature vectors (multidimensional numerical data) for processing by advanced machine learning techniques.
7. Remote operation of robots and autonomous systems through natural language – Helen Hastie, Xingkun Liu (Heriot-Watt University)
Natural language provides an intuitive, hands-free means of communicating with robots and autonomous systems. This interaction can help the operator understand the vehicle activity e.g. “Where are you? What are you doing?”, but can also help manage missions and give commands: “Go and inspect the north tower and send back images”. We will demonstrate an interactive system that is an extension of the MIRIAM interface with this enhanced functionality.
8. MIRIAM – Multimodal Intelligent inteRactIon for Autonomous systeMs – David Robb (Heriot-Watt University)
The goal of ORCA Hub is that through the use of robotic systems offshore, the need for personnel in hazardous environments will decrease. It is, however, essential that onshore operators maintain situation awareness in order to monitor the mission and handle unforeseen circumstances that may affect their intended behaviour, such as a change in the environment. We present a multimodal interface called MIRIAM (Multimodal Intelligent Interaction for Autonomous systems) that combines visual indicators of status with a conversational agent component. MIRIAM offers a fluid and natural way for operators to gain information on vehicle status and faults, get explanations of system behaviour, as well as, mission progress and set reminders. This interaction type leads to increased transparency, which will facilitate a trusting relationship. This is key for adoption of robotic systems, in particularly key in high-stakes, hazardous situations.
9. Probabilistic model checking for robots deployed in extreme environments – Xingyu Zhao (Heriot-Watt University)
Robots are increasingly used to carry out critical missions in extreme environments that are hazardous for humans. This requires a high degree of operational autonomy under uncertain conditions, and poses new challenges for assuring the robot’s safety and reliability. The team have developed a framework for probabilistic model checking to verify the safety and reliability requirements of such robots, both at pre-mission stage and during runtime.
10. Towards certification of unmanned aircraft for offshore asset inspection – Matt Webster, Vince Page, Charles Patchett, Mike Jump (University of Liverpool)
In future inspections of offshore assets utilising robots (including unmanned aircraft, submersibles and ground-based vehicles), they will not only be expected to collate new data from their payload of instruments, but will also be expected to interact with the infrastructure being inspected, undertake remedial tasks and engage with embedded monitoring systems of the asset. This increasing level of interaction and deployment frequency of robot inspections requires an understanding of how we can embed safe and trusted operational architectures within robots.
Currently, robots can undertake constrained semi-autonomous inspections, using predetermined tasks (missions) with minimum supervision. However, the challenge is that the state of the world changes with time as does the condition of the robot. Therefore, robots must be able to undertake adaptive measures to support optimal outcomes during autonomous missions. In this presentation, the team will show an initial architecture to the safe verification and validation of health condition and certification of robotic and autonomous inspection systems for offshore assets.
11. Tether-powered aerial drone demonstration – Chang Liu, Arnau Garriga Casanovas (Imperial College London)
Imperial College London’s Chang Liu demonstrated the automatic mapping process of a tether-powered aerial robot using an automatic reel system for tether management during flight.