Throughout history, technological advancements have improved the way in which we operate. The industrial revolution replaced human labour with machines, dramatically improving the efficiency of human labour while creating entirely new products and service offerings. Artificial Intelligence (AI) has the potential to have a similar effect, by replacing or augmenting the human mind, to allow us to work more effectively or to offer entirely new capabilities.

AI is a broad category of techniques that allows machines to mimic the capabilities of the human mind. IBM, which is implementing a variety of these techniques across its products, has published a good definition on what AI means to us today. Some of the current applications include

  • Natural Language Processing: Which mimics the capability to understand speech and text
  • Machine Learning: Which mimics the capability to learn from experiences and data
  • Computer Vision: Which mimics the ability to recognise images and classify objects
  • Data Mining: Which mimics the ability to discover patterns in large data sets.

There has become an increasingly ubiquitous application of AI techniques across a variety of industries in recent years. A major benefit driving this increase in utilisation, is that AI can remove human from unfavourable operational environments. Professor Robin Murphy categorised these unfavourable operating environments in his book, Introduction to AI Robotics, as falling into one of three categories of Dull, Dirty and Dangerous. These categories out outlined in the graphic below.

Figure 1. Unfavourable operating environments for human performers

 

An example of the emerging use of AI to industry automation can be found in the rail industry. Drawing inspiration from Rio Tinto’s use of driverless heavy-freight trains in remote areas of Australia, now rail operators in the USA are testing new systems for autonomous operation of freight trains through towns and cities. These systems will incorporate networks of sensors, including machine vision systems for detecting broken wheels and to perform automated track inspection.

There are other examples of these technologies being developed on Australian shores too. The Boeing Australia developed Loyal Wingman is an advanced unmanned aerial vehicle (UAV) being procured by the Royal Australian Air Force. The UAV uses AI to enable independent flying (without an operator) to perform reconnaissance and early warning missions, while maintaining safe distances from manned aircraft.

While the development and realisation of these technologies is exciting, they also create new challenges that need to be addressed. These include new potential failures not previously experienced by other technologies, the unpredictability of AI due to the non-deterministic and evolving behaviour and performance and the lack of trust in the system and its interactions with its operating environment.

In our forthcoming series of articles, Shoal will investigate ‘The AI Revolution’. We will explore the challenges in adopting these technologies and possible steps required to ensure the technology integrates with the existing environment, providing beneficial outcomes to society.  We will look at technology trends, current examples, and future technologies, across industry, and their strategic implications.

Follow us on LinkedIn to keep up-to-date and join the discussion and connect with our authors, Jake Vanderlinde and Ben Mashford.