A four-legged robot that can complete a marathon on a single battery charge has been demonstrated by a team from KAIST. The robot, which completed the Sangju Marathon in South Korea in 4 hours and 19 minutes alongside human runners, traveled three times as far per charge than existing robots and could pave the way for improved battery life in legged robots.

Legged robots have the potential to assist in rescue operations in mountainous or disaster-stricken areas. However, recent advances in quadruped robots have been limited by their ability to operate beyond approximately 20 kilometers (12 miles) because of limited battery life. Unlike robots on wheels, legged robots must continuously expend energy to support their body weight and compensate for energy lost with each ‘step,’ reducing the distance they can travel between charges.

In their article published in Nature, Jemin Hwangbo and colleagues present RAIBO2, a four-legged robot designed to reduce energy loss through both software and hardware. Its features include lightweight legs, optimized motor-driving circuits and movement rules reinforced by machine learning.

In the marathon, the robot maintained an average speed of 2.64 meters per second (5.9 miles per hour), navigating a notoriously difficult course with 286 meters (938 feet) of elevation change and slippery sections of terrain. Upon completion of the run, RAIBO2 had used 1,280 Watt-hours (Wh) of energy and achieved an average total cost of transport of 0.25 (a measure of the energy expended during motion, factoring in the weight of the robot).

The authors propose that this is the first quadruped robot to have a lower cost of transport than the human value of 0.37. Based on the robot’s total battery capacity of 2,016 Wh, the authors estimate that RAIBO2 could have traveled a further 25 kilometers (16 miles). The robot travels approximately three times as far per battery charge as existing quadruped robots.

The findings offer insights into improving the efficiency of legged robots, with the aim of extending their operational ranges and boosting their utility.

Who’s behind this story?

Lisa Lock

BA art history, MA material culture. Former museum editor, paramedic, and transplant coordinator. Editing for Science X since 2021.

Full profile →

Robert Egan

Bachelor’s in mathematical biology, Master’s in creative writing. Well-traveled with unique perspectives on science and language.

Full profile →

How many times have you asked someone to hold open a door while carrying something, or fumbled with your front-door keys with your arms full of groceries? Wouldn’t it be great if you could give yourself the extra hand you need?

This is one of the goals of human movement augmentation, a new and growing area of research that aims to extend human capabilities beyond their natural limits. Within this field, researchers are developing and studying artificial limbs that can be added to the human body. These are known as supernumerary robotic limbs, or SRLs.

Current SRLs are largely experimental, but in the future wearable robotic limbs could help in a range of areas. They could have an impact in manufacturing, health care, space, music, arts, gaming and rehabilitation.

Unlike prostheses, SRLs increase the overall capacity of the human body rather than restoring the capacity of a missing limb. If you’re familiar with the Marvel universe, think of Doctor Octopus—his mechanical arms are fictional examples of this type of technology.

In the real world, scientists have created extra arms a little like Doc Ock’s (but benign), as well as additional legs and fingers.

These extra limbs don’t look or behave like actual human arms or legs. They’re combinations of rigid materials, motors and actuators—mechanical devices that produce force. They can be worn like a backpack or watch.

So far, they tend to be controlled using another body part, most often the foot. By using pedals or shoes fitted with sensors, a person can generate forces or make movements with their feet that are then translated into movements of a robotic limb. Recent research has also investigated using the muscles around the ears for control.

Designers typically avoid using the arms and hands as controllers. This is to maximize the added capacity additional limbs provide.

A common question is whether robotic limbs could be controlled by the brain. They aren’t yet, though scientists have explored the idea.

When the brain sends electrical signals through the nerves to control muscles, only certain frequencies of those signals actually convey information that instructs movement. There is essentially spare bandwidth in the body’s nervous signaling network.

Some researchers have tried to see if this spare capacity can be harnessed—independently of actual muscle movement and noninvasively—to control SRLs.

Although promising, this approach has had a low success rate. Its complexity and the long calibration and training period needed for users probably make it unsuitable.

Two arms good, four arms better?

Supernumerary robotic limbs are designed to provide something additional to the natural body. They should work alongside our natural limbs without limiting their natural functions. The goal isn’t to create autonomous entities that can do a job in place of someone’s body. Rather, it’s to give people additional tools they can control and use.

For example, you might one day wear a backpack with two robotic arms attached and use these to keep a ceiling panel in place while you screw it to the ceiling. Or you could have a belt with two extra legs to keep you stable in an unnatural position, leaving your hands free to fix something. Or you could have a sixth finger, which could allow you to open a bottle with one hand.

These examples might seem like science fiction, but they’ve all been realized and tested in the lab.

Industrial uses are beginning to be explored. A future surgeon might operate four surgical tools at the same time, for instance. Or a rescue operator could immediately extract an earthquake victim while lifting debris, without needing to wait for support.

But for now, such uses remain ambitions. This field is still largely experimental, and the performance of SRLs is limited. Existing devices are often heavy, bulky and difficult to wear, and their speed is usually limited for safety.

Learning to use them well also takes a lot of effort, especially as current control interfaces tend to be complex and unintuitive. That said, studies of how people respond to using supernumerary robotic limbs have found that they can learn the basics of controlling a third arm or sixth finger in less than an hour.

Feedback plays a key role in this learning process. Vibrating motors and electrodes that provide a small current can help people “feel” a robotic limb they’re controlling.

This is crucial for motor learning: When you try to play tennis or the piano, your haptic senses play an important role in allowing your brain to learn the correct movement. The same applies with a robotic limb. Sensory feedback is also essential for getting the brain to treat the SRL like it is part of the natural body.

Despite the challenges of learning how to use them, research shows that people become more adept at using extra robotic limbs over time, and that the brain seems to adapt and incorporate these extra limbs to the point that having them feels natural. This suggests there’s real future potential in the field.

In the meantime, it’s important to raise awareness about human movement augmentation and SRLs specifically. Most people have no idea what human augmentation means or that these devices exist.

Now that you know that they do, what would you use these limbs for?

Who’s behind this story?

Gaby Clark

MA in English, copy editor since 2021 with experience in higher education and health content. Dedicated to trustworthy science news.

Full profile →

Robert Egan

Bachelor’s in mathematical biology, Master’s in creative writing. Well-traveled with unique perspectives on science and language.

Full profile →

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Mechanical and aerospace engineering professor Anupam Pandey’s research team has found that microscopic folds in soft materials can act like tiny, programmable traffic gates for liquid droplets. The folds can stop droplets, let them pass or merge them together without surface coatings or additional hardware. Remarkably, the droplets never touch the fold. They sense it from a distance and slow down, stop short or change course in response. The study is published in the Proceedings of the National Academy of Sciences.

The researchers worked with “creases”—narrow, self-contacting downward folds that form when a soft, rubber-like surface is compressed. They found that these creases curve the surface around them, creating an energy barrier a droplet feels from a distance. Droplets smaller than a critical size stop before touching the crease, while larger droplets pass through, creating a sharp threshold.

Adjusting how much the surface is compressed changes the size cutoff, allowing the same crease to be reprogrammed on the fly to block or admit different droplets. The effect is disproportionate: Squeezing 15% harder quadruples the threshold size.

“What we did not expect is that drops of different sizes are not sensing the same thing,” says Pandey, senior author of the study. “The large ones respond to how steep the fold is, the small ones to how quickly that steepness changes.”

Building on that basic gating effect, the team showed that creases can be arranged to perform more complex tasks: guiding droplets along set paths, sorting them by size or surface tension, storing a kind of droplet “memory” that keeps track of past inputs, reshaping a stream of droplets into fewer, larger pulses, and even carrying out logic operations similar to those in electronic circuits.

In one demonstration, the team routed two streams of droplets into a single crease to build a half adder, the arithmetic unit at the base of every processor. All of these tasks were accomplished using nothing but the mechanical state of the surface itself.

“Because the crease appears and disappears with compression, the circuit is rewritable,” Pandey says. “We can switch a gate off, let everything through, and switch it back on. Nothing is permanently patterned into the surface, and the control comes down to a single mechanical variable.”

Because the surfaces are simple to make and require no batteries, motors or embedded circuitry, the researchers say the approach could be useful for portable diagnostic devices that analyze small fluid samples and for systems that harvest water from fog or humid air.

Co-authors on the paper, “Creases gate and steer droplets via elastocapillary repulsion,” are Zixuan Wu, a postdoctoral researcher in Pandey’s group; Gavin Linton, an undergraduate in mechanical and aerospace engineering; and Stefan Karpitschka, a professor of physics at the University of Konstanz in Germany.

Who’s behind this story?

Robert Egan

Bachelor’s in mathematical biology, Master’s in creative writing. Well-traveled with unique perspectives on science and language.

Full profile →

When people hear the term “green building,” they often imagine expensive technologies, costly developments or environmentally friendly features that only wealthy buyers can afford.

But green buildings do not have to be luxury log homes surrounded by lush vegetation. A green building can be any size. It is designed, built or upgraded to use water, energy and other resources in a way that causes less harm to the environment.

For example, a green building would use natural light and ventilation to reduce the need for electric lighting, air conditioning and fans. It could be built with more sustainable materials, like recycled steel, reclaimed wood, locally made bricks or low-carbon concrete.

Indoor spaces in green buildings are designed to be comfortable and reduce running costs over the building’s lifetime. Good insulation, energy-efficient lighting, solar panels and systems that collect rainwater can achieve this.

We are sustainable housing researchers who’ve spent years researching how green housing has evolved. In a recent study, we tracked the growth of green building certification in South Africa. We looked at 510 buildings that received green certificates between 2011 and 2023.

Our analysis included where they were located, who owned them and the ratings they received. We also looked into whether the certification system stops developers from making misleading claims about how green a building is.

Our research found that green building is growing and spreading across South Africa. However, it is mainly happening in Johannesburg, Cape Town and Sandton. This is largely through upgrades to existing buildings rather than new green buildings, showing that green feature upgrades add value to older properties.

Most buildings received only midlevel green ratings, however. The highest rating for green buildings—six stars—is quite rare.

Some buildings received very low ratings, which means that “greenwashing” (pretending to be green) might be happening.

The next step is to establish how much value green upgrades and certification can add to South Africa’s existing properties.

How buildings get a green rating

It costs different amounts to get certified, depending on the building. A green building rating measures how well a building saves energy, water and other resources, and reduces carbon emissions and environmental harm.

South Africa’s Green Star system, introduced in 2008, rates new and existing buildings, interiors and neighborhood developments. This system was developed by the Green Building Council South Africa. It sets a common standard for developers, investors, financiers, owners and occupants to judge how environmentally friendly buildings are.

In South Africa, it’s not compulsory for new buildings to be green. It’s up to investors, developers, building owners and users to opt for green building or not. This is similar to the situation in many developing countries, where including green features in housing developments remains voluntary rather than compulsory.

But voluntary does not mean unimportant. Green buildings are increasingly attractive because they can reduce resource consumption and improve energy and water efficiency.

They also appeal to buyers for their quality and sustainability. Green buildings can have improved indoor air quality, lighting and other features. Passive design can improve them, too, through the direction a building faces, the size and placement of its windows, and the use of suitable building materials to make the most of natural light and airflow while keeping indoor temperatures comfortable.

Without rating systems and certification, it can be difficult for consumers and investors to tell if a building genuinely reduces its environmental impact or is just claiming to do so.

Green building certification is growing rapidly in South Africa

Our study found a 1,694% jump in the number of certified green buildings in South Africa, from about 50 to over 897 between 2014 and 2024. This suggests increasing market interest in sustainable buildings. Green development is becoming more established within the property sector.

The banking and commercial property investment sectors are actively adopting green building practices in South Africa, we found, driven by strong financial, retail and office markets. Educational institutions are also becoming early adopters. There’s increasing demand for green buildings in student residences and higher education facilities.

For developers and property owners, certification can offer several advantages, including increased rental and capital value, as tenants, investors and buyers put sustainability first when deciding where to live.

Certification can also help property owners qualify for green financing, especially with banks and development finance institutions investing more in environmentally responsible projects.

The problems with green certificates

Apart from benefits, there are also challenges. One concern is greenwashing, in which buildings are marketed as environmentally friendly without delivering meaningful environmental improvements.

Greenwashing happens when developers exaggerate a building’s environmental benefits to attract buyers, investors and tenants.

Our study found warning signs of this. Most certified buildings met the four-star “best practice” standard, but six-star ratings—awarded for “world leadership”—were rare. Some existing buildings also received ratings below four stars.

As certification becomes more popular, building owners, investors and regulators need to check that it reflects real improvements in energy use, water use and environmental impact, and is not just being used as a marketing tool.

What needs to happen next

Encouragingly, we found that green building certification is starting to spread beyond South Africa’s largest cities. This could make greener building practices more practical and accessible in other parts of the country.

Builders, developers and homeowners may still wonder whether going green is affordable and worthwhile. A useful first step is to ask a suitably qualified building professional how much green features would cost and how much they could save over the building’s lifetime. These features can be included in a new design or added to an existing building.

Green certification could help reduce energy and water use, improve building practices and make it easier to qualify for green financing. But it must lead to real environmental improvements. As certification grows, government, investors and buyers need to ask: Are these buildings genuinely becoming greener, or do they simply have a green label?

Who’s behind this story?

Gaby Clark

MA in English, copy editor since 2021 with experience in higher education and health content. Dedicated to trustworthy science news.

Full profile →

Andrew Zinin

Master’s in physics with research experience. Long-time science news enthusiast. Plays key role in Science X’s editorial success.

Full profile →

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Modern data centers, high-precision measurement systems and high-resolution radar sensors rely on extremely broadband amplifier chips that combine low noise with high output power. Researchers at the Fraunhofer Institute for Applied Solid State Physics IAF have developed a monolithic microwave integrated circuit (MMIC) with strong performance in these areas.

The distributed amplifier (DA) achieves a gain of 11 ± 2 dB at frequencies from 4 to 420 GHz. It is based on high-electron-mobility transistors (HEMTs) fabricated in the indium gallium arsenide (InGaAs) on silicon (InGaAs-on-Si) material system with a gate length of 20 nm. Previous generations of the chips were fabricated on gallium arsenide (GaAs) substrates with a gate length of 35 nm.

The MMIC was first presented in an article published last year in IEEE Microwave and Wireless Technology Letters.

High bandwidth with high output power and low noise

“The MMIC amplifier we developed is characterized by a unique combination of the relevant parameters: bandwidth, noise and output power. Up to a frequency of 202 GHz, it achieves noise figures between 3.2 and 7.7 dB; in a setting optimized for low noise, these range from 2.8 to 6.7 dB. The saturated output power ranges between 6 and 8 dBm,” explains Dr. Fabian Thome, developer of the MMIC and deputy head of the High-Frequency Electronics Business Unit at Fraunhofer IAF.

Broadband MMICs in practice: Data centers, measurement systems, radar sensors

Such high-bandwidth amplifiers are particularly relevant for the latest generation of data centers that use optical data transmission. They are used in interfaces that convert optical signals to electrical signals or electrical signals to optical signals. Optical data transmission enables higher speeds and more compact designs. The increasing prevalence of artificial intelligence applications is driving a sharp rise in demand for modern data centers.

In addition, high-bandwidth chips with low noise and high output power enable more precise measurement systems and higher-resolution radar sensors. Higher-performance chips increase the resolution, range and sensitivity of these systems.

At European Microwave Week (EuMW 2026), taking place in London from October 4 to 9, 2026, Fraunhofer IAF researchers will present exhibition samples of the amplifier chip as well as other innovations in high-frequency electronics at Booth B35. FormFactor, Inc., a collaborating measurement technology manufacturer, will demonstrate a live measurement of the chip at its booth (D10).

Who’s behind this story?

Lisa Lock

BA art history, MA material culture. Former museum editor, paramedic, and transplant coordinator. Editing for Science X since 2021.

Full profile →

Andrew Zinin

Master’s in physics with research experience. Long-time science news enthusiast. Plays key role in Science X’s editorial success.

Full profile →