This year's iteration of the RoboCup officially kicked off on Thursday in Sydney, which is the first time it has been held in Australia since 2000. RoboCup is an annual robotics competition that promotes robotics and AI research by pitting different universities and research organisations against each other in various robotic challenges.
Held at Sydney's International Convention Centre, RoboCup 2019 is running 13 competitions across four days, ranging from robot soccer to robots rescuing victims placed in a simulated field.
The robots participating this year have come a long way from when the RoboCup first began, University of Technology Sydney (UTS) RoboCup team leader Mary Anne Williams said.
Williams attended her first RoboCup back in 2002 when robots could only contain 16MB of memory. Compared to the robots used in RoboCup 2019, such as Softbank's NAO Robot, the processing power of robot sensors have improved radically where they can now perform tasks such as finding an object in a room and moving it to a designated area.
For this year's RoboCup, UTS is competing in the RoboCup@Home League event, where robots perform actions that assist human life in domestic environments, such as picking up the garbage from a kitchen bin.
The UTS RoboCup team has placed second at this event in the past two RoboCups.
While Williams hopes UTS can bring home the trophy this year, she told ZDNet that the aim of the competition isn't necessarily about winning but about building resilience in robots so they can be used by humans to improve their livelihoods.
"[Our robot] doesn't die and crash when the network goes down, which is fairly diabolical in this setting where you've got a robot at home. Because to bring a robot back to life requires a bit of technical expertise and your average home dweller probably doesn't have that expertise so it's super important to have that robot running even if it's being repaired remotely," Williams said.
The UTS professor also believes competitions such as RoboCup are extremely valuable for students -- whether they are at university, high school, or primary school -- as it allows them to stand out in Australia's tech workforce as well as learn how to work together in creating innovation.
"It's not [just] about robots, big companies come to RoboCup to recruit, and the reason is that roboticists are the cream of the crop of data scientists. The reason for this is, of course, robots are a data collection machine; it's collecting vision, sound, touch, a whole bunch of multi-modal data which it then has to integrate, make sense of, and then you know, make decisions in real-time," Williams said.
"Our robots make about 30 decisions a second, so if you can build a software system like that then everybody wants you. Most students don't actually go into robotics industry, the robotics industry is not big enough to take on everybody yet, they find places like Uber, Apple, Twitter, and Google."
The AI developed by UTS for the RoboCup@Home League event was a three-year project by the university's social robotics research team, and the project will be wrapped up after the end of this year's RoboCup.
The UTS social robotics team is currently working with the Southwest Sydney local health district on various robotics applications that are expected to kick off next year. They have plans to begin building prototypes for these applications in the second half of this year.
The university in September last year also opened a AU$65 million research facility that focuses on exploring areas such as civil and environmental engineering, mechanical and mechatronic engineering, software, and electrical and data engineering, to provide new services to industry.
Around 170 teams and 1,200 participants from universities and research organisations across the globe are expected to compete at this year's RoboCup.
The finals for the RoboCup 2019 events are on Sunday.
It uses laptops to collect data on students' hand gestures, eye movements through webcams, as well as mouse, keyboard, and digital pen movements to better understand the link between behaviour patterns and learning outcomes.