Robotics, Digital Twins and AI: Connecting the Dot Matrix

Abstract

Abstract Robotics is often referred to as an enabler towards safer and more cost-effective operations within the energy industry. However, for robots to achieve their full potential they too require their own enablers. This paper intends to present a collection of lessons learnt in robotic development to date that has resulted in the necessity to develop an encompassing digital architecture. This architecture, with an artificial intelligence component, has been designed to optimise both robot and digital twin capability, where the complete system is always working with the latest information available that reflects the ground truth. The architecture concept will be explained including why integration and standardisation are of paramount importance. The paper will also demonstrate how business value can be generated before fully explainable, transparent, verifiable autonomy is available for robotics to be deployed on future unmanned platforms that have been designed to optimise robot operation opposed to existing human engineered environments. All these topics are important within their own right, but all are also years away from adoption on an industrial scale. To ensure continued engagement until such a point is reached, the industry needs to focus more on the quick wins and immediate value that can be gained from robotics. The impact of COVID-19 has shown how quickly and effectively the energy industry has been able to transition to remote working. This represents a massive opportunity to prove what robotics can deliver via remote operation, minimizing vendors where possible. The concept of "full" autonomy will be discussed with respect to value generation and barrier to entry or adoption within the energy industry. The paper will address all practical considerations that have required attention to date plus explain the next programme of work in the evolving RAS (robotics and autonomous systems) digital architecture whilst ensuring complete integration with the data pipelines that have already been built.