Hamburg-News-media-it-creative-industries-innovation-future-business
Roboter-Fußball-Team der TUHH - © Team HULKs

Humanoid Robotics: Step by step towards normality

Will robots soon be planning our days? Research into the future is being conducted playfully way at the Hamburg University of Technology (TUHH). Part 5 of our AI SERIES

Are robots guided by artificial intelligence (AI) now part of our everyday life or do they still belong to the world of science fiction? Samsung, for instance, presented three different service robots (the bots Care, Air and Retail) at the CES 2019 electronics trade fair in Las Vegas in mid January. They can be used variously for assisting the elderly, testing and improving air quality, and operate as a kind of butler in restaurants and hotels. A sleep robot (Somnox) that is supposed to help with difficulty in falling asleep and a robot that folds the laundry (Foldimate) were also exhibited at the CES. This shows that there is international research in many directions.

HULKs: Grasping AI through robot football

The Hamburg University of Technology (TUHH) is working on the humanoid robotics of the future. “We are looking into how robots can be used initially in games, but later in real life,” said Patrick Göttsch, a 35-year-old postgraduate student at the TUHH and chairman of the HULKs student-working group. The Hamburg Ultra Legendary Kickers are a robot football team consisting of knee-high NAOs – humanoid robots from the French manufacturer SoftBank Robotics – that have autonomously mastered passing the ball, one-on-one contests and tactics using advanced AI. Serious research lies behind this gaming approach. Researchers from all over the world test their skills against each other at the RoboCup robot football competition. Business partners including Eppendorf Instrumente GmbH, freiheit.com, sensor producer Sick and Ibeo Automotive Systems are backing Hamburg’s young scientists.

Robots using pixels to analyse surroundings

The researchers hope, by means of the football setting, to attract maximum attention, with a view to introducing as many people as possible to AI. The intention is that around the year 2050robots will be able to beat the ruling human world champions. “Opinion on whether this is realistic is divided,” said Göttsch. “There are wild optimists and pessimistic doubters.” A lot remains to be done. While a healthy adult is able to order and manage their environment quite naturally, a robot has to be programmed step by step. “Our world presents itself to a robot in the form of colour data – pixels – from which it generates abstract information, on the basis of which algorithms go to work. AI controls motion and strength, for instance, with the aim of achieving motion processes that are flowing, robust and as rapid as possible,” Göttsch explained.

Humanoide Robotik: Schritt für Schritt zur Normalität
Patrick Göttsch

Robots as home assistants

The use of AI is currently being trialled in various areas, whether in self-driving cars, drones or household robots like vacuum cleaning and mopping robots, window cleaners and gardener robots. Most of these robots are small and compact flat-shaped devices. The coming generations of household or service robots may turn up in humanoid form, “as this increases their acceptability,” Göttsch believes. The use of humanoid robots is of particular interest in the care sector, given demographic shifts and an ageing society, and could help people to stay in their homes longer. Personal robots in the home could help remember medication schedules and call for help, if their owner has fallen.

Super intelligence not yet on the horizon

A basic distinction needs to be made between strong and weak AI. “Strong AI” or “super intelligence” that is equivalent to or even surpasses human intellectual capabilities remains a vision for the distant future. By contrast, “weak” or “narrow” AI aims at fulfilling well-defined tasks and is already in use in many areas of everyday life, such as in navigation systems, automated translation and image and speech recognition. “The AI of today is very good at identifying recurring data in structures from a large quantity of information and evaluating them from differing points of view,” Göttsch said. This opens up a multiplicity of applications in fields like medicine, traffic management and predictive planning – the use of predictive models based on statistical methods and machine learning to arrive at reliable predictions of business planning.

Transition eventually leading to AI

In this, the use of AI today is directed at helping humans, not at pushing us out. “AI is intended purely to manage the data flow and to back up doctors or traffic planners e.g. with the results or to enable robots to take on work that requires exceptional strength or that humans dislike.” Nevertheless, Göttsch certainly believes that we are in a transitional phase, in the course of which more and more areas will be taken over by automation, AI and robots.

Rethinking basis of our livelihoods

And what will become of humans? If gainful employment falls away, and people are no longer able to provide for themselves and their families, this will lead to existential difficulties. For this reason, a societal and economic transformation process must accompany the technological transformation process, Göttsch believes. “We will have to rethink the basis of our livelihoods.” One option: If paid work disappears, working for the common good could be rewarded. “Newly-gained liberties could allow people to develop their creativity in the arts, socially or in research, depending on their skills and talents.”
ys/rm/pb

Source and further information:
www.tuhh.de
www.hulks.de

Read the other parts of our AI series as well:
Part 1: Artificial intelligence – a tool not a mind
Part 2: jung diagnostics: Algorithms for MRI image analysis
Part 3: My colleague, the robot – a popular member of staff?
Part 4: Artifical intelligence – catalyst of positive future

More articles

  • There are no older articles in this category.
  • There are no newer articles in this category.