March 4, 2020

Psychological theory could help teach robots to walk

By Ellen Daniel

A psychological theory known as perceptual control theory could help accelerate the important but challenging task of teaching robots to walk.

A study by the University of Manchester, published in the Journal of Intelligent and Robotic Systems, brings together researchers from the field of clinical psychology and robotics engineering to investigate self-balancing robots.

Teaching a robot how to move on two legs without falling over is a deceptively complex task, and robots that “walk” in an inverted pendulum, similar to humans, require balance and agility.

How perceptual control theory can help robots walk

The study was based around perceptual control theory. This is a model for how organisms relate to the world around them based on the idea that responses to the external environment is varied based on perception.

Perceptual control theory has been used in psychological therapies, but has never been applied to robotics before. It is hoped that it could lead to improvements in the way robots walk.

Researchers studied what happened when the standard algorithms controlling a simple droid, which moves on similar to how humans walk on two legs, were replaced by algorithms based on perceptual control theory. They trialled robots programmed with three different software controllers: proportional control, LQR and perceptual control theory.

In the study, the robot using perceptual control theory was able to balanced more “accurately, more promptly and more efficiently” by assessing its environment 100 times a second.

This allowed the robot to balance in similar way to how a human stay upright when walking, meaning it could move more effectively, whereas the other robot was less stable and wobbled more.

Computer technologist Dr Rupert Young said: “This research is a peek at a radical new way of understanding how to build robotics systems, that are dynamic and adaptive despite the chaotic, unpredictable nature of the real world. Based upon an elegant and natural approach, this paradigm holds the promise of developing far more sophisticated, autonomous robots.”

Read more: Industrial exoskeleton sees Hyundai workers wear robots to combat fatigue.

Verdict deals analysis methodology

This analysis considers only announced and completed cloud-deals deals from the GlobalData financial deals database and excludes all terminated and rumoured deals. Country and industry are defined according to the headquarters and dominant industry of the target firm. The term ‘acquisition’ refers to both completed deals and those in the bidding stage.

GlobalData tracks real-time data concerning all merger and acquisition, private equity/venture capital and asset transaction activity around the world from thousands of company websites and other reliable sources.

More in-depth reports and analysis on all reported deals are available for subscribers to GlobalData’s deals database.

Topics in this article: ,