The study, which leader Professor Rob Riener said was the first to comprehensively compare humanoid robots with humans, was carried out at ETH Zurich in Switzerland.
The work was partially inspired by cinema – in the Terminator sci-fi films, for example, the eponymous robots are so ‘perfect’ that they are often superior to humans. But, the researchers asked, “how well do humanoid robots perform nowadays away from the cinema screen?”
Professor Riener set out to answer that question by first developing criteria to enable a “meaningful comparison” between humans and machines. An industrial robot painting car bodies on a production line can do it faster, for longer and more precisely than a human, for example, but it is specially developed for just that task and does not have any other abilities, so those types of machines were excluded.
“We humans shape our environment according to our criteria and needs. If robots are to support us in a meaningful way, they need to work in this manmade environment. We therefore quickly arrived at robots that are similar to humans, at least anatomically,” said Professor Riener, a specialist in sensory-motor systems and founder of the Cybathlon project, focused on assistive technologies.
The study therefore examined exclusively humanoid robots, with 27 relevant examples selected.
The researchers also defined certain selection criteria within that category. “For example, for a robot that has rollers rather than legs, it would be fairly easy to roll faster than a human can run – but we didn’t want to compare apples with pears,” said Professor Riener.
To address this, robots were only compared if they had two or four legs and were able to climb steps. They also needed a slim figure to pass through doors, a certain height (at least 50 cm), and arms and hands so that they could pick up objects on a tray or shelf. To be able to work with and support humans, they also needed to be quiet and not give off any exhaust emissions.
Initial results surprised even Professor Riener – comparing the individual ‘components’ of machines and humans, such as microphones with ears, cameras with eyes, or drive systems with muscles, the technical components always fared better in ‘key sensory-motor properties’.
Carbon fibre is often used in robots, for example, which is harder than bone. “If we disregard other properties of the human bone, such as the fact that it is self-healing, the technical solution is clearly superior in terms of mechanical features,” the researchers said.
Faced with the superiority of individual parts, Professor Riener explained the “baffling” issue that comes up: “The question that arises is why we are not able today to construct a robot from these high-quality components that has better powers of movement and perception than humans.”
The second result of the study showed that that has yet to happen – “if we consider the activities that humans and machines are asked to carry out, humans are generally superior to robots,” the ETH announcement said.
Although humanoid robots are able to walk and run, most machines are unable to keep pace if the walking or running speed is set in relation to body dimensions, weight or energy consumption. At 6.1 metres per second, MIT’s robot Cheetah can run faster than a jogging human, for example, but it has a high energy consumption (973 watts) and is also only deployed under laboratory conditions. Humans also “significantly” outperform robots in terms of endurance versus operating time, the study found.
Robots benefit from high precision in certain scenarios. “For example, when balancing on one leg, robots can easily stiffen their joints, while humans tend to wobble a bit – which costs considerably more energy. Robots can also precisely recognise their joint angles and repeat movements very accurately, which is pretty impressive,” said Professor Riener.
The results were more mixed around picking up objects. While the robots could pick up objects extremely quickly, they were not able to match humans’ many different hand movements and the manipulative skills of biological fingers.
Robots were also only able to perform some types of movement, such as swimming, crawling and jumping, while most humans are capable of performing and combining several of these movements. Playing football was cited as an example – machines are still a long way from dribbling, heading, or analysing and interpreting the strategy of other players.
Humanoid robots are not just a gimmick however, Professor Riener said: “The progress made by robotics in recent years is incredible. We wish to have robots around us so that they can help us with difficult or dangerous tasks.
“However, our manmade environments are very complex, and it’s therefore not so easy for robots to function here autonomously and without error. Nevertheless, I’m confident that with the powerful technical components that are available we will soon be able to construct more intelligent robots that are capable of interacting with us humans better.”
System engineering and automatic control technology will enable better combinations of the powerful components, the researcher said. Deployment might then be conceivable in nursing homes, construction, or in the household, the announcement added – wherever support is urgently needed to relieve staff and support people with limited mobility.
The work was published today (7 November) in Frontiers in Robotics and AI.
Want the best engineering stories delivered straight to your inbox? The Professional Engineering newsletter gives you vital updates on the most cutting-edge engineering and exciting new job opportunities. To sign up, click here.
Content published by Professional Engineering does not necessarily represent the views of the Institution of Mechanical Engineers.