QUT launches robot with better vision and human smarts

QUT launches robot with better vision and human smarts

Future lies in robots that are intuitive and come equipped with reflexes similar to humans, says Queensland University of Technology

Baxter comes equipped with vision

Baxter comes equipped with vision

Robots that are perceptive and work with better reflexes alongside humans are the next big thing, according to a world-renowned roboticist at the Queensland University of Technology (QUT).

Professor Peter Corke, director of the Australian Centre for Robotic Vision (ACRV) at QUT Science and Engineering Faculty, said these advancements in robotics help improve the vision and perception of robots.

He said it enables robots to work more intelligently alongside humans, together with faster reflexes, response time and understanding the terrain.

The QUT, a global front-runner in robotics research and development, has unveiled a new robot called Baxter that does just that. This assistant works alongside researchers in a safer and highly-interactive environment.

The more intuitive “seeing and perceiving” features have been built into Baxter, enabling it to perceive the world or environment as humans do, with better reflexes.

"Once robots can see and understand the environment they operate in they can make decisions that allow them to work safely beside us," said Professor Corke.

Working with humans

He noted that current industrial robots are dangerous for humans to work around. This is largely because they're simply not equipped to recognise and avoid obstacles suddenly in their way.

"Baxter uses a range of sensors to detect movement around it, as well as spring-loaded joints which stop it continuing a pre-programmed movement if it meets an unexpected obstacle,” said Professor Corke.

Robotic vision is a key technology that enables robotics to transform labour-intensive industries. The more intuitive thinking machines will become a ubiquitous feature of the modern world, according to the Australian Centre for Robotic Vision's chief operating officer, Dr Sue Keay.

"We consider that 'seeing' is far more than just processing images,” she said. "It's a complex process tightly coupled to both memory and action, which gives robots the understanding they need to robustly perform tasks that involve objects and places while at the same time providing rapid and continuous feedback for control.”

Robots that can see, learn and respond as humans do will increase productivity across key sectors and industries, Dr Keay said.

Professor Arun Sharma, the QUT’s deputy vice-chancellor (Research and Commercialisation) said a new generation of innovating robotic vision experts are converting ground-breaking research into new products, services and enterprises.

In a rush

The QUT’s research team has tried to “rush” Baxter several times. This was to assess its reflexes, as well as quicker responses and turnaround times. These responses replicate how humans interact in a multi-dimensional environment.

Baxter was also tested to use computer vision, while testing its reflexes to play an unbeatable game of Connect Four. This robot is now being trained to recognise and pick a colour-coded vegetable.

The university’s robotics’ centre galvanises some of Australia’s top computer vision and robotics researchers. These groups are banding to create the next generation of robots that can see, understand and respond to the more complex, real-world environments.

Among the recent initiatives, teams are researching the use of robotic vision to create advanced farming tools. Robotics technology will help tackle Crown of Thorn starfish outbreaks on the Great Barrier Reef.

The QUT is part of the Australian Research Council Centre of Excellence – a coalition that brings the University of Adelaide, Australian National University, Monash University, NICTA, as well as other partnering international universities.

Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.

Join the newsletter!


Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.

Tags robotsQueensland University of TechnologyAustralian Centre for Robotic Vision (ACRV)Australian Research Council Centre of ExcellenceRobotic vision

More about Australian National UniversityAustralian Research CouncilCrownMonash UniversityNICTAQueensland University of TechnologyQueensland University of Technology (QUT)TechnologyUniversity of Adelaide

Show Comments