Dr Rachel Kallen – Dr Michael Richardson | Stepping into the Future: Enhancing Interactions Between Humans and Machines
About this episode
Over the last few decades, technology has become an integral part of our lives. Although society has largely embraced this leap, many people don’t believe that machines can perform on an equal level to humans. Dr Michael Richardson and Dr Rachel Kallen at Macquarie University in Sydney explore how machines can be developed to interact naturally and effectively with humans. Read More
Our human behaviours often involve coordinating our actions with others. Working with others is more efficient and depends upon our skills in social perception. Dr Richardson and Dr Kallen acknowledge that for artificial agents to be accepted within society, they need to have real-time coordination and the ability to respond in a ‘human-like’ way.
Much of their research has focused on testing the effectiveness of dynamic motor primitives, a mathematical formulation of human perceptual-motor behaviour, for controlling the actions of artificial agents.
To assess the effectiveness of artificial agents in collaborative tasks, the researchers conducted a study comparing the performance of human-to-human and human-to-machine pairings. In the experimental task, the pairs worked together to herd a flock of virtual sheep. Interestingly, the human-to-machine performance was the same as that of the human-to-human pairings.
The results also showed that complex human movement and social behaviours can be successfully implemented by artificial agents, especially when multiple parties are required to work together.
A further study aimed to assess whether artificial agents could perform ‘pick and place’ activities. We undertake these types of actions daily, such as setting the table or loading the dishwasher. The research found that dynamic motor primitives were successful in instructing artificial agents when to pass or not pass an object, and which hand movements to make.
Another aim of this study was to determine whether humans were able to detect the involvement of a machine. The experimental set-up involved participants wearing virtual reality headsets that masked whether their partner was a human or a machine. They worked together to move coloured discs from one area to a corresponding target. By the end of the task, the participants were unable to tell whether they had been working with a human or a machine.
Dr Richardson and Dr Kallen also explored the application of artificial agents in training humans, particularly for tasks requiring motor behaviours. To be successful, artificial trainers would need to exhibit natural human behaviour.
The team’s study used another virtual sheep herding task to assess whether machines could provide a comparable level of training to human experts. The results demonstrated that humans could learn successfully from a machine and were also able to use this to enhance their future performance with other human team members.
Dr Richardson and Dr Kallen have demonstrated the critical role that artificial agents can play in the training of motor perception skills across a range of industries, including healthcare and sports. Over time, machines will play an increasing role in improving our collaborative work, and have many benefits to offer society.
Original Article Reference
Summary of the papers: ‘A comparison of dynamical perceptual-motor primitives and deep reinforcement learning for human-artificial agent training systems’, doi.org/10.1177/15553434221092930; and ‘Human social motor solutions for human-machine interaction in dynamical task contexts’, doi.org/10.1073/pnas.1813164116
This research is supported by the Australian Research Council and the Australian Department of Defence, Science and Technology Group
This work is licensed under a Creative Commons Attribution 4.0 International License.
What does this mean?
Share: You can copy and redistribute the material in any medium or format
Adapt: You can change, and build upon the material for any purpose, even commercially.
Credit: You must give appropriate credit, provide a link to the license, and indicate if changes were made.
Are you ready to increase the impact of your research?
Stay Up To Date With SciTube
Subscribe to receive our latest videos straight to your mailbox