The distinction between automated and autonomous systems is a critical and often misunderstood concept, particularly within the drilling industry. We address this ambiguity in this paper by developing a comprehensive framework to classify the level of autonomy in drilling systems.
While current taxonomies provide a basis for classification, they often fall short of capturing the nuances of true autonomy. To overcome these limitations, we first synthesize definitions of autonomy from various high-risk disciplines, including aerospace and control theory, to identify key traits. These traits, such as environment complexity, problem complexity, mission, risk management, dynamic planning, situational awareness, decision-making, execution, learn from experience, and interaction with agents, are then used to build a novel framework tailored to the specific demands of drilling operations.
Our proposed framework introduces quantifications for each of these traits. By encoding these traits, our framework allows for a systematic and objective evaluation of a system’s autonomy. This approach challenges the notion that autonomy requires the complete absence of human interaction. Instead, it recognizes that an autonomous system can collaborate with human operators while retaining the independent authority to make and execute decisions.
This new framework provides a clear and consistent method for assessing autonomy, facilitating the development and deployment of truly intelligent drilling technologies.
This abstract is taken from paper SPE 217754 by J.P. de Wardt, de Wardt and Co.; E. Cayeux and R. Mihai, NORCE; J. Macpherson, Baker Hughes; P. Annaiyappa, Independent Consultant; and D. Pirovolou, Weatherford.The paper has been peer reviewed and is available as Open Access in SPE Journal on OnePetro.