Robots can’t feel; these sensors could change that

Penn State

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

video: 

Researchers have proposed new design for sensors that allow for teleoperated control of robotic and prosthetic systems. As seen in the demonstration, a robotic hand powered by the team’s sensors can accurately track the gestures and movements of a human hand. To ensure the robot doesn’t damage held items, the array's design allows the hand to be highly sensitive when grabbing or holding delicate objects.

view more 

Credit: Provided by Larry Cheng / Penn State

UNIVERSITY PARK, Pa. — A research team, including Huanyu “Larry” Cheng, James L. Henderson Jr. Memorial Associate Professor of Engineering Science and Mechanics at Penn State, is using pressure sensors — tiny devices, roughly the size of a paperclip, that can measure the force applied over an area — to design a highly sensitive electronic “skin” to use alongside robots and prosthetic limbs.

Cheng is a corresponding author on a paper, recently published in Nano-Micro Letters, that introduces the improved pressure sensor design. The team’s sensors can be assembled into an interconnected array, offering researchers and clinicians a wireless approach to recognizing spatial pressure distribution, hand gestures and even different types of food based off their weight and texture.

In the following Q&A, Cheng discussed pressure sensing technology and how his team’s work could help robots accurately “feel” the sensation of touch.

Q: Why do current pressure-sensing technologies struggle to balance sensitivity and accuracy? How did you address these issues with your new design?

Cheng: It is still difficult for flexible pressure sensors to simultaneously achieve high precision and responsiveness to subtle pressures, despite extensive research and development. Conventional designs often provide abundant conductive networks, but their irregular arrangement weakens compressive strength, which limits detection range and long-term stability.

In this work, we designed a flexible pressure-sensing platform based on a material known as reduced graphene oxide aerogel (rGOA) — an incredibly lightweight, oxygen-rich material. Using freeze casting, a manufacturing technique that solidifies mixtures of liquids and solids into one material, we can form our sensors to have an anisotropic microstructure, meaning they have different mechanical strengths depending on the direction we apply stress.

With these adjustments, our sensors can simultaneously achieve ultrahigh sensitivity, a broad pressure detection range and long-term stability. Although a single sensor is only about eight millimeters in size, they can each support about three ounces of force and reliably load and unload weight over 20,000 times. By assembling individual sensors into an interconnected array, we can effectively create an artificial “skin” capable of precisely measuring extremely subtle changes in pressure.

Q: How are the sensors built? How did you test their effectiveness?

Cheng: The pressure sensor was fabricated by sandwiching rGOA between a synthetic, plastic-like film stamped with interdigital electrodes — small measurement devices printed onto the material in silver ink — and a layer of thin, silicon-based polymer material. Sandwiching the materials together ensures stable electrical contact, mechanical robustness and flexibility for practical applications.

We tested our sensors by measuring the current response under a wide range of applied pressures, while also assessing frequency response and stability under a range of temperatures and humidities. Our sensors proved extremely sensitive, offering almost twice as much sensitivity as sensors manufactured with traditional structures. Additionally, the sensors exhibited incredibly fast response and recovery times, responding to pressure changes in just over 100 milliseconds, and recovering from responses in only 40 milliseconds — a process that other sensor options can take over 250 milliseconds to fully cycle through.

Q: What does the process of assembling the sensors into an “artificial skin” look like? What sorts of devices and applications could use such a skin?

Cheng: The sensors can be assembled into an array, collecting many individual measurements. Using a microcontroller, a tiny computer designed to execute a specific task, these pressure signals are collected, converted into digital values and visualized in real time. This allows the sensors to identify the position and magnitude of pressure caused by different objects, which can be helpful for prosthetics, robotic manipulation and battery health monitoring.

The sensors’ flexibility, ultrahigh sensitivity and environmental stability let the arrays conform to complex surfaces for precise pressure mapping, a way to visualize how much pressure is between two surfaces. These capabilities open new possibilities in smart robotics, wearables and human-machine interfaces, enabling the detection of dynamic pressure changes from irregular objects or small volume shifts. One key application is early identification of battery swelling in electric vehicles — a common issue where rising internal pressure can cause irreparable damage to a battery.

Additionally, the sensors can identify object shapes or help robotic systems grasp fragile objects. When used alongside a robotic manipulator like a hand or a vice, the sensors can monitor pressure in real time and compare it with preset safety thresholds to prevent object damage. This force-feedback system allows the robot to accurately track hand movements and grasp delicate objects such as tofu, cotton and steamed buns, which could be a big step forward in effective human-machine interaction and interfacing.

Q: What’s next for this work? Are there plans for commercialization in the future?

Cheng: We plan to reduce the sensor size and weight to enhance biocompatibility and stability in complex environments. Additional research could enable spatially programmable sensitivity — allowing a single sensor or array to simultaneously detect subtle pressures in one region and withstand large loads in another — while also integrating pressure, temperature or strain sensing within a single, comprehensive structure.

We believe these sensors have a strong potential for future real-world deployment and commercialization through integration with wearable devices and commercial robots. Arrays of these sensors could offer a low-cost and high-performance sensing solution, while remaining highly flexible and customizable.

Collaborator and funding details can be found in the paper. The team has filed a provisional patent for this technology.

At Penn State, researchers are solving real problems that impact the health, safety and quality of life of people across the commonwealth, the nation and around the world.    

For decades, federal support for research has fueled innovation that makes our country safer, our industries more competitive and our economy stronger. Recent federal funding cuts threaten this progress.  

Learn more about the implications of federal funding cuts to our future at Research or Regress.

 

The sensors can visualize the applied pressure of objects with varying weights and shapes in real time, offering researchers an in depth look at how different household items interface with the sensors.

Credit

Provided by Larry Cheng / Penn State

---

Journal

Nano-Micro Letters

DOI

10.1007/s40820-026-02109-8 

Method of Research

Experimental study

Subject of Research

Not applicable

Article Title

Graphene Aerogel-Based Flexible Pressure Sensor for Physiological Signal Detection and Human–Machine Interaction

Article Publication Date

27-Mar-2026

 

Researchers identify potential disease marker, therapeutic target for cats with osteoarthritis

]North Carolina State University

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

By comparing osteoarthritis pain pathways known to be active in dogs and humans to those in cats with degenerative joint disease (DJD), researchers found that elevation of a particular molecule, artemin, could serve as a marker of disease (and possibly pain) as well as a potential therapeutic target. The findings offer the most comprehensive evidence to date that naturally occurring osteoarthritis (OA) in cats mirrors important biological features of human disease.

DJD, the main form of which is osteoarthritis, affects humans and all animals, including cats. However, the mechanisms underlying DJD-associated pain in cats are poorly understood.

In humans and dogs, transient receptor potential (TRP) ion channels – cellular sensors for a wide spectrum of physical and chemical stimuli that are found in the spinal cord’s dorsal root ganglia (DRG) – are activated to express osteoarthritis pain. One mechanism by which TRP ion channels can be activated is by a molecule called artemin. When artemin binds to its receptor, GFRA-3, it sends a signal to activate TRP channels, leading to a series of events which register the pain.

“We know that Artemin/GFRA-3 induced changes to TRP channels play a role in osteoarthritis pain in humans and dogs, but didn’t know whether cats shared this biological pathway,” says Santosh Mishra, associate professor of neurobiology at North Carolina State University and co-corresponding author of the study.

“With relatively limited options for managing chronic pain in cats, if we can understand more about how the sensation of pain is being generated in cats themselves, then we will be a step closer to developing therapies that are effective in cats,” adds Duncan Lascelles, professor of translational pain research at NC State. Lascelles is a co-corresponding author of the study.

The research team, comprised of experts in integrated molecular neuroscience, clinical pain phenotyping and translational methodology, designed an interdisciplinary study that yielded a robust data set.

First, they did detailed pain and DJD status assessments on more than 70 cats so that they could accurately compare samples from healthy and DJD cats. Then they looked at blood serum and DRG tissue samples from both healthy cats and cats with DJD.

The researchers confirmed that the TRP channels commonly associated with osteoarthritis pain in humans and dogs were expressed and functional in healthy cat DRG neurons. Then they compared expression of the TRP channels and GFRA-3 in healthy cats to those with DJD. Finally, they compared artemin concentrations in blood samples from healthy cats and cats with DJD.

They found that the TRP pathways associated with pain were active in healthy and DJD cats, and that increased artemin blood concentrations were correlated with radiographic or X-ray confirmation of DJD, but not necessarily with pain.

“These results are interesting for a couple of reasons,” Mishra says. “First, we now know that the biological pathways are similar, but the differences are also important.

“We saw that artemin levels were increased in cats with radiographic evidence of the disease, but that artemin levels didn’t correlate to veterinarian-assessed pain. However, knowing how difficult it is to measure pain in cats, in future work we will utilize technology to more objectively measure pain.”

The researchers also point to the elevated artemin in blood serum as a biological marker of osteoarthritis in cats as well as a potential therapeutic target.

“If veterinarians could do a blood test for increased artemin to diagnose DJD instead of X-rays it would save time and stress for cats,” Mishra says. “And perhaps targeting artemin expression could be therapy for either pain or disease progression. Now that we know the pathway is conserved, we can dig deeper into the mechanisms to find therapies.”

“Because cats exhibit naturally occurring DJD/OA similar to people, this work provides a valuable window into real biological processes and pain mechanisms which will ultimately improve clinical care for cats,” says Lascelles. “The findings may also help refine translational models and inspire cross‑species therapeutic advances.”

The study appears in Frontiers in Pain Research and was supported by the EveryCat Foundation (formerly the Winn Feline Foundation) under grant W18-028 and the National Institutes of Heatlth/National Institute of Arthritis and Musculoskeletal and Skin Diseases under award 079713. Joshua Wheeler, postdoctoral student at NC State, is first author. Other NC State collaborators include Margaret Gruen, professor of clinical sciences; and former postdoctoral student Chie Mochizuki, who is currently at the University of Tennessee.

-peake-

Note to editors: An abstract follows.

“Artemin/GFRA3 Axis and TRP channels: Molecular Insights from a Feline Model of Osteoarthritis”

DOI: 10.3389/fpain.2026.1716651

Authors: Joshua Wheeler, Margaret Gruen, Duncan Lascelles, Santosh Mishra, North Carolina State University; Chie Mochizuki, North Carolina State University and the University of Tennessee
Published: March 16, 2026 in Frontiers in Pain Research

Abstract:
Introduction: Degenerative Joint Disease (DJD) is a form of highly prevalent osteoarthritis in humans and animals, including cats, which causes significant pain and hypersensitivity. Despite its prevalence, the mechanisms underlying the DJD-associated pain in cats are poorly understood. While transient receptor potential (TRP) ion channels are expressed in the dorsal root ganglia (DRG) and are implicated in osteoarthritis pain (e.g., through Artemin/GFRA3-mediated changes to TRPV1 and TRPA1 electrical properties), there is currently only indirect evidence of TRP ion channel expression in the feline DRG. This study aims to address this knowledge gap.
Methods: Fura-2 based in vitro calcium imaging was used to confirm the functional expression of TRPV1, TRPV2, TRPA1, and TRPM8 in healthy cat DRG neurons. A quantitative reverse transcription-polymerase chain reaction (qRT-PCR) with SYBR green was used to confirm and compare mRNA expression of pain sensors including TRPV1, TRPV2, TRPV4, TRPA1, TRPM3, TRPM8, MRGPRD, TAC1, and GFRA3 in the DRG neurons of healthy cats and DJD group. Finally, serum artemin concentrations were quantified using enzyme linked-immunosorbent assay (ELISA) and compared between healthy and DJD cats.
Results: Functional responses of TRPV1, TRPV2, TRPA1 and TRPM8 were determined via calcium imaging in DRG neurons obtained from healthy cats. Gene expression is further extended into healthy versus DJD cats. While TRPV1, TRPV2, and TRPM8 showed a >1.5-fold increase in cats with DJD compared to healthy controls, MRGPRD mRNA expression showed a corresponding ∼1.5-fold decrease. However, these increases or decreases in fold-change did not reach statistical significance. GFRA3, a receptor for artemin, was found to be expressed in the cat DRG, though its levels remained unchanged in DJD-affected cats. Lastly, a significant association was found between serum artemin concentrations and radiographic DJD scores but not with veterinarian pain scores.
Discussion: Our findings characterize functional expression of several pain and hypersensitivity-associated TRP ion channels in cat DRG neurons and identify the Artemin/GFRA3/TRP axis as a potential driver of chronic pain. The expression of channels, including TRPV1, TRPA1, and TRPM8 modulated by Artemin/GFRA3 pathway was confirmed. Bridging these cellular findings to DJD state, the observed correlation between serum artemin concentrations and radiographic DJD scores further implicates this pathway in disease severity. These results provide potential early evidence that the Artemin/GFRA3/TRP axis drives pain in feline DJD. This conservation is consistent with findings in other species, such as rodents and canines, suggesting translational relevance for therapeutic targeting.

---

Journal

Frontiers in Pain Research

DOI

10.3389/fpain.2026.1716651 

Method of Research

Experimental study

Subject of Research

Animals

Article Title

“Artemin/GFRA3 Axis and TRP channels: Molecular Insights from a Feline Model of Osteoarthritis”

Article Publication Date

17-Mar-2026

 

Improved catalyst enhances the conversion of ethanol to hydrogen

Fine-tuning the processing of perovskite-type ceramic material increases system stability and reduces costs by eliminating the need for noble metals.

Fundação de Amparo à Pesquisa do Estado de São Paulo

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

High-resolution microscopy showing ruthenium exsolution in LaCrO3 perovskite. The process is similar to nickel exsolution 

view more 

Credit: Fabio Coral Fonseca

Amid the climate crisis and the urgent need to reduce greenhouse gas emissions, hydrogen has emerged as one of the most promising energy sources for the transition to a low-carbon economy. When produced from renewable sources, it can serve as a clean fuel, strategic industrial input, and means of energy storage.

In the Brazilian context, hydrogen production from ethanol, especially when derived from biomass, is a particularly promising pathway. Brazil has a well-established infrastructure for producing, distributing, and using this biofuel, opening the door to technological solutions that add value to ethanol and expand its role in the energy transition.

A study led by Fabio Coral Fonseca, a senior researcher at the Institute of Energy and Nuclear Research (IPEN), made significant progress in this direction. The study demonstrated that fine-tuning the processing of a perovskite-type ceramic catalyst is crucial for maximizing the conversion of ethanol into hydrogen. This increases the stability of the system and reduces costs. It also eliminates the need for noble metals that are traditionally used in this type of reaction. The study was published in the International Journal of Hydrogen Energy.

The conversion is carried out through a process known as ethanol steam reforming (ESR). In simple terms, this involves reacting ethanol with steam at high temperatures to produce hydrogen and carbon dioxide. The ideal overall reaction, which maximizes hydrogen production, can be represented as follows: C₂H₅OH + 3 H₂O → 2 CO₂ + 6 H₂. However, in practice, the process involves several intermediate steps. This makes the role of the catalyst central to directing the reaction, maximizing hydrogen yield, and avoiding undesirable pathways. One such pathway is the formation of coke, or carbon deposits, which rapidly degrade the material.

“Catalysis is a surface property. What we want are very small particles that are very well distributed and stable over time,” says Fonseca. “The problem is that, at high temperatures, these particles tend to shift, agglomerate, and lose activity.”

To address this challenge, the study used a perovskite-type ceramic oxide. Unlike conventional catalysts, however, the active element in the reaction, nickel (Ni), is not impregnated onto the surface of the ceramic; rather, it is incorporated into the crystalline structure of the material during synthesis. “Rather than placing the metal on top of the support, as in classical catalytic methods, we introduce the nickel into the crystal structure. Then, under controlled conditions, that nickel emerges on the surface,” the researcher explains.

This phenomenon, known as “exsolution,” causes metallic nickel nanoparticles (Ni⁰) to emerge on the surface of the solid. They are strongly anchored to the substrate, which gives them much greater stability against sintering and carbon deposition. “The metal comes from the inside out. It doesn’t move around the surface as it does in impregnated catalysts. That gives the system much greater stability,” says Fonseca.

The central breakthrough of the study was demonstrating that a seemingly simple parameter – the calcination temperature of the precursor oxide prior to the reduction step – controls the entire performance of the catalyst. The researchers synthesized the material chemically and calcined it at three different temperatures: 650 °C, 800 °C, and 1,200 °C. This step, which precedes the actual catalytic reaction, determines the microstructure of the solid, particularly the size of the ceramic particles and the available surface area.

“If we heat the perovskite to very high temperatures, it grows too much. And that hinders the exsolution of nickel later on,” Fonseca notes. The results showed that calcination at 650 °C preserves a larger surface area, whereas higher temperatures promote grain coalescence, drastically reducing this area. Smaller ceramic particles favor nickel exsolution and the formation of smaller, more active nanoparticles. “The key point of the study was to show that substrate size controls exsolution. If the particles are large, the nickel doesn’t exsolve well. If they’re smaller, it exsolves more efficiently,” the researcher summarizes.

In ethanol steam reforming tests, the catalyst that was calcined at 650 °C produced significant results: 100% ethanol conversion, a yield of 4.04 moles of H₂ per mole of ethanol, and stable operation for up to 85 hours with low coke formation. In contrast, materials calcined at 800 °C and 1200 °C exhibited lower nickel exsolution, lower conversion, and a shift in reaction selectivity, favoring the simple dehydrogenation of ethanol over complete reforming for hydrogen production. “It isn’t enough to choose the right elements. How the material is manufactured is decisive. A relatively simple adjustment in processing completely changes the performance,” emphasizes Fonseca.

The researcher places the study within a broader technological agenda. According to Fonseca, converting ethanol into hydrogen is not always the best solution from an energy standpoint, especially when considering mobility. “Ethanol is a very valuable molecule. To obtain it, you have to go through agriculture, fermentation, and distillation. Simply breaking it down to produce hydrogen and then electricity may not be the best choice,” he notes. For this reason, the group is investigating direct ethanol fuel cells, which can convert the liquid fuel directly into electricity. “Ultimately, we study these perovskites because they fit very well with that technology,” Fonseca adds.

Perovskites are materials defined less by their specific chemical composition and more by their characteristic ABO₃ crystalline structure. This structure, first observed in the natural mineral calcium titanate (CaTiO₃) – the original perovskite – is now reproduced synthetically and in various ways in the laboratory. In this architecture, various elements can occupy the A and B sites of the crystal lattice, giving these materials extraordinary structural flexibility to tailor their electrical, ionic, magnetic, and catalytic properties.

The work on nickel is part of a broader strategy to explore metal exsolution in perovskites. In a previous study conducted with groups in the United States as part of a project supported by FAPESP and the National Science Foundation (NSF), the IPEN team obtained significant results with ruthenium exsolved from lanthanum chromite (LaCrO₃)-based perovskites. In this case, ruthenium – an even more active metal in reforming reactions – is initially incorporated into the crystal lattice. During the ethanol reforming reaction, it emerges as metal nanoparticles that are strongly anchored to the support. This study was also coordinated by Fonseca and published in the journal Catalysis Science & Technology.

Studies with polycrystalline powders, as described in the two articles, are just one part of the IPEN scientists’ research program. The team is transitioning to more controlled systems based on epitaxial thin films produced by pulsed laser deposition. “Epitaxial” refers to material that grows in an ordered manner on top of another material, copying its crystalline structure. “In this case, what we do is compact the powder, transform it into a ceramic wafer, and then sublimate that material with a high-energy laser. The vapor deposits onto a well-ordered substrate and forms a nearly perfect crystal,” Fonseca describes. This approach enables the study of exsolution at the atomic level using advanced characterization techniques at Sirius, the Brazilian synchrotron light source.

By demonstrating that high catalytic performance can be achieved with abundant, low-cost metals offering strong stability, the studies point to a concrete path toward reducing dependence on noble metals and making sustainable hydrogen production more viable. In the Brazilian context, where ethanol is abundant and demand for low-carbon energy solutions is growing, these results reinforce the potential of the ethanol-hydrogen route. More broadly, they reinforce the potential of exsolved perovskites as a strategic resource for the energy transition.

Support for the nickel study came from FAPESP through the Thematic Project “Advanced Electrochemical Devices for Molecular Conversion and Energy Production”, Research Grants 17/11937-4, 18/19251-7, and 24/00989-7, as well as a Doctoral Scholarship.

About São Paulo Research Foundation (FAPESP)
The São Paulo Research Foundation (FAPESP) is a public institution with the mission of supporting scientific research in all fields of knowledge by awarding scholarships, fellowships and grants to investigators linked with higher education and research institutions in the State of São Paulo, Brazil. FAPESP is aware that the very best research can only be done by working with the best researchers internationally. Therefore, it has established partnerships with funding agencies, higher education, private companies, and research organizations in other countries known for the quality of their research and has been encouraging scientists funded by its grants to further develop their international collaboration. You can learn more about FAPESP at www.fapesp.br/en and visit FAPESP news agency at www.agencia.fapesp.br/en to keep updated with the latest scientific breakthroughs FAPESP helps achieve through its many programs, awards and research centers. You may also subscribe to FAPESP news agency at http://agencia.fapesp.br/subscribe.

---

Journal

International Journal of Hydrogen Energy

DOI

10.1016/j.ijhydene.2025.153326 

Article Title

Calcination temperature of the perovskite parent compound controls active metal exsolution and catalytic performance for ethanol steam reforming

 

Two organs, one brain area: How fish orientate themselves in the water

A brain region integrates input from the eyes and the ‘third eye’ in fish to help them orientate themselves

Osaka Metropolitan University

Facebook

XLinkedInWeChatBlueskyMessageWhatsAppEmail

 

image: 

Light is detected by both the eye and the pineal organ. The light-sensitive opsin PP1 in the pineal cells senses the balance of ultraviolet and visible light and converts it into neural signals. These signals are processed in the tegmentum, where they regulate the fish’s up and down swimming behavior.

view more 

Credit: Osaka Metropolitan University

Using zebrafish, researchers from Osaka Metropolitan University (OMU) have identified the tegmentum region in the fish midbrain as the area where light input from both the fish’s eyes and the pineal organ—the ‘third eye’—is integrated. Their findings suggest that fish use the integrated light signals in this region to swim up or down in response to differences in the wavelength of light.

In the aquatic world, light changes depending on depth, water conditions, and differences such as sunlight and shade. Differences in the levels of visible and UV light enable fish to infer these factors, which they may use to make survival decisions.

To understand the related processes taking place in the brain, an OMU research team led by Professors Akihisa Terakita and Mitsumasa Koyanagi with Dr. Seiji Wada of the Graduate School of Science looked at the opsin parapinopsin 1 (PP1). Opsins are specialized proteins that respond to light. They are typically found in the eyes, but in some species, opsins like PP1 are also found in the pineal organ. Using calcium imaging, the team investigated how color-detection signals produced by PP1 in the pineal photoreceptor cells are passed to the brain by nerve cells.

“We decided to study zebrafish, as their larvae are transparent,” Professor Koyanagi said. “This transparency means that changes in calcium levels within nerve cells can be observed as changes in the fluorescence intensity of the calcium indicator, allowing us to measure the strength of neural activity.”

PP1 exhibits opposite responses to UV and visible light. Using calcium imaging, the group traced these responses to light from the pineal organ to the tegmentum via ganglion cells.

“Our study showed that the tegmentum integrates visual information from the eyes that is combined with color information detected by the pineal organ. These integrated signals then contribute to the fish’s up and down swimming behavior,” Dr. Wada, the first author of the paper, said.

When they raised fish without the PP1 gene, they did not show the typical responses to changes in the wavelength of light.

“These findings shed light on how animals process visual information, advance the analysis of neural circuits using light, and expand research into behavioral control,” Professor Terakita said. “In the future, these findings may contribute to applications in neuroscience and biomedicine, such as the identification of neural circuits using PP1-based optogenetics.”

###

About OMU 

Established in Osaka as one of the largest public universities in Japan, Osaka Metropolitan University is committed to shaping the future of society through “Convergence of Knowledge” and the promotion of world-class research. For more research news, visit https://www.omu.ac.jp/en/ and follow us on social media: X, Facebook, Instagram, LinkedIn.

---

Journal

Proceedings of the National Academy of Sciences

DOI

10.1073/pnas.2520290123 

Method of Research

Experimental study

Subject of Research

Animals

Article Title

Neural circuits for decision making based on pineal photoreception in zebrafish

Fish change their swimming direction in response to differences in the wavelength of light: when ultraviolet light is weak, they swim toward the surface, whereas when it is strong, they swim downward.

Credit

Osaka Metropolitan University