Skip to main content

Automation, Trust and Workload

Principal Investigator

Andreas Duenser, Senior Research Scientist (email)

CRP Team Members

Martin Lochner, Postdoctoral Fellow, (former CSIRO Team member), UTAS (collaborator)

Andrew Heathcote, Professor of Cognitive Psychology, UTAS (collaborator)

Timothy Gale, Senior Lecturer Engineering, UTAS (collaborator)

Benjamin Brooks, Associate Professor Seafaring, AMC / UTAS (collaborator)

Luke Strickland, Postdoctoral Fellow, Cognitive Psychology, UTAS (collaborator)

Kristy de Salas, Senior Lecturer in Computing and Info Systems, UTAS (collaborator)

Christopher Lueg, Professor of Computing, UTAS (collaborator)

Mark Billinghurst, Professor of Human Computer Interaction School of Information Technology and Mathematical Sciences, UNISA (collaborator)

Other collaborators

Andy Reeson, Principal Research Scientist and Team Leader

Claire Mason, Experimental Scientist

Lifetime

2016 – ongoing

Description

This is a Collaborative Research Project with collaboration partners form the University of Tasmania and the University of South Australia and applications in transportation, health and FinTech

Automation and autonomous systems are playing an increasing role in every-day life. Whereas the previous generation of autonomous technology had its impact largely in industrial spaces, (e.g. automotive factory robots), current advances are bringing such systems into the homes, workplaces, and automobiles of today’s citizens. A key factor here is the increased proximity to human operators and bystanders. This project aims to develop a useful model for making predictive analyses based on the three factors of: Automation, Trust, and Workload.

Our framework includes physical automation (e.g. self-driving cars, autonomous robotics) as well as automation within a user-experience framework, considering automated online information systems. Recent research on human interactions with autonomous and decision support systems indicates that the level of trust in the system affects the success of the interaction on a number of levels, and should therefore be taken into account as a major factor in the design process. Therefore the relation between trust, autonomy, and workload is intrinsic to the human-machine system, and the development of a model surrounding these influences will inform the successful future development of such systems.

Projects

Automatic blood-oxygen targeting in neonatal intensive care – workload and trust in automation

This project is a collaboration with our UTAS CRP members (led by Timothy Gale). The goal is to investigate the effect of automating blood-oxygen targeting in neonatal intensive care in terms on  nurse’s workload and trust in this automation. Initially this will be studied in a simulated neonatal intensive care environment before moving into a real environment.

Trust in Automated Financial Advice

This project was part of the CSIRO-Monash Superannuation Research Cluster. Out team collaborated with the Inclusive Socio-Technical Innovation Team team to study people’s trust and motivation in using automated financial advice for superannuation planning.

An increasing range of financial services have been automated over the last couple of decades.  Financial advice is the new frontier for automation, with a number of ‘robo-advisor’ products under development or beginning to interact with customers.  Such automated systems have great potential to cost-effectively extend financial advice beyond the minority who currently enjoy it.  However, it faces barriers, both in the form of many people’s low engagement with superannuation, as well as their wariness of trusting technology with their finances.

We designed and conducted two studies: one online survey-based study to better understand people’s perception of automated advice, and an experimental online survey to study attitudes towards the automation of financial advice, and how experience with a simple automated financial advice system impacts users’ attitudes, motivation, and trust.

Managing Vessels in Port Waters

This project is a collaboration with Australian Maritime College (UTAS) and the CSIRO CI team to design a system to aid the navigation of large vessels as they enter or leave a port.

AMC_6104

The overall objective of the project is to identify mechanisms for managing vessels in port waters that will improve productivity, occupational health and safety (OHS), and operations from a safety point of view.

 

AMC custom ‘full mission’ maritime simulator main bridge, in Launceston, Australia
Figures 1 and 2:   AMC custom ‘full mission’ maritime simulator main bridge, in Launceston, Australia

Our focus in this project is:

  • Developing and testing new measurement tools of team workload in real-world and simulated environments (including psycho-physiological measurements).
  • Developing human performance testing protocols and performing system evaluations

Given the safety-critical nature of operations on a large ocean going vessel, and considering the need to understand operator workload for individuals in the distributed maritime operations team (including Captain, Pilot, Tug Master(s), and Vessel Traffic Service), we have begun investigating the remote collection of Electro-Dermal Activity (also known as Galvanic Skin Response) as a correlate of mental workload.  Electrophysiological signals are collected via personal radio transmitters for each team member, and these measures are combined with more traditional methodology (Instantaneous Situation Awareness measure; Subjective Workload Assessment Technique; Communications Analysis) in order to develop an understanding of the mental state of each team member during both standard operation conditions, and during critical emergency events such as engine failure, steering failure, and communications failure.

 

Mine Informatics project

The CSIRO Mine Informatics project is working with the Mining Industry to address productivity improvements by the application of intelligent informatics methods. The main activities are:

  • mapping and characterising major voids in underground mines using remotely operated flying vehicles; and
  • virtual modelling of mining and mineral processing operations, enabling automation of the discovery of opportunities for improving efficiency through the application of machine learning techniques.

Cognitive Informatics team members help address the second of these challenges, helping to develop interfaces and visualisations to make intelligent optimisation comprehensible to a wide range of users within mines, such as mine managers and supervisors, planners, maintenance controllers and procurers.

mine_informatics
Figure 1: Operations centre.