Research News

Tiny robot team could be a gamechanger for safety inspections

Jeff Zehnder — Wed, 21 May 2025 20:25:11 +0000

Tiny robot team could be a gamechanger for safety inspections Jeff Zehnder Wed, 05/21/2025 - 14:25

One slithers. One crawls. Neither looks like much on their own. But together, they form a super team—one that might just change how we inspect the most complicated machines in the world.

Kaushik Jayaram, an assistant professor in the Paul M. Rady Department of Mechanical Engineering at 91PORN, is working to build the next generation of robot inspection tools by studying some of nature’s simplest creatures.

This robotic duo is about as odd as it is ingenious: tiny, insect-inspired robots paired with inflatable vine-like robots that grow like plants and curl like snakes. These high-tech helpers can navigate a complex maze of machinery and squeeze through the tightest of spaces—like the guts of a jet engine—to potentially perform non-destructive evaluation faster, cheaper and better than ever before.

The tiny mCLARI robot, developed by Assistant Professor Kaushik Jayaram and his team in the Animal Inspired Movement and Robotics Laboratory.

“If you look at the infrastructure around us, there are a lot of buildings, bridges, dams and machines that have all of these little nooks and crannies,” said Jayaram, who is also affiliated with the BioFrontiers Institute, the Biomedical Engineering Program, the Robotics Program and the Materials Science and Engineering Program. “They need very careful, regular inspection and maintenance, but there’s just no easy, cost-effective way to get in.”

Jayaram said there is also an element of public safety involved. According to the Federal Aviation Administration, nearly 15% of aviation accidents are caused by mechanical malfunction.

In just this year alone, the National Transportation Safety Board has reported 94 aviation accidents, 13 of which have been identified as fatal incidents.

“When it comes to tasks such as flying, where human safety is paramount, we need aircraft technology and machinery to work 100% of the time,” Jayaram said. “Our research is one of the efforts to address these concerns using the advantages of robotics.”

The work, in collaboration with Laura Blumenschein at Purdue University, has drawn interest from the U.S. Air Force Research Laboratory. They’ve awarded the two researchers a three-year, $1.4 million grant to prove these small robots can work together to produce big results.

But as unlikely as this robotic team might seem, Jayaram believes they have the perfect blend of “offense” and “defense” to get these dirty and delicate jobs done.

First on the roster is Jayaram’s mCLARI microrobot. This tiny machine—weighing in at less than a gram—can climb, squeeze through cracks the size of a penny and move with a millimeter precision.

However, due to its small stature, it struggles to carry any extra weight. Large batteries and electronics are incompatible with the little robot, and without them it cannot travel long distances or maneuver tight spaces effectively.

The inflatable vine-like robot, developed by Laura Blumenschein, an assistant professor at Purdue University.

That’s where its vine-like teammate comes in. This robot can inflate like a party favor, allowing it to carry more weight and conform to the environment. In Jayaram’s vision, the inflatable snake can act as mCLARI’s personal Uber driver, negotiating constraints of tight spaces and dropping the tiny robot directly at the site of inspection.

Once in location, Jayaram said the mCLARI robot, fitted with cameras and miniature evaluation sensors, can gather and transmit real-time data for offline analysis. When it’s done, it can hop right back on the snake-like robot and the team can make the winding journey back home, saving hours of evaluation time and thousands of dollars in service costs in the process.

“Each of the robotic systems have their own pros and cons,” said Jayaram. “By combining the strengths of these two robots, we’re overcoming the disadvantages to create a single collaborative system that can give us quick insight into these compact and confined spaces.”

But this tiny squad of robots is capable of much more than just inspection. In fact, Jayaram dreams of a day where his insect and vine-inspired robotic friends can be deployed in a variety of scenarios where being small, agile and adaptive are a premium.

Maybe one day this robotic team can play a vital role in environmental monitoring to detect high-risk wildfire zones and prevent ecological damage. Or maybe they can be used in disaster response situations—like a collapsed building—to help save human lives.

Jayaram said the possibilities are truly endless.

“These small, confined crevices and spaces are actually way more ubiquitous than we originally thought. Even in the medical arena—if we shrink these robots even further, make them shapeshift, and use biocompatible materials, maybe our technology can one day be crawling inside our bodies, detecting and releasing blood clots or taking measurements just like a pill,” Jayaram said. “We get very excited when we think about the future. If we can build systems that can effectively navigate the world and combine them with sensors, we can do a lot.”

Off

Traditional

White

Fleeting fireflies illuminate Colorado summer nights—and researchers are watching

Jeff Zehnder — Fri, 02 May 2025 18:29:21 +0000

Fleeting fireflies illuminate Colorado summer nights—and researchers are watching Jeff Zehnder Fri, 05/02/2025 - 12:29

New research uses firefly flashing patterns to identify species and what they’re communicating. Read from CU experts Orit Peleg and Owen Martin on The Conversation.

Off

Traditional

White

Engineer nabs prestigious grants to design insect-inspired, shapeshifting robots

Jeff Zehnder — Tue, 29 Apr 2025 14:56:02 +0000

Engineer nabs prestigious grants to design insect-inspired, shapeshifting robots Jeff Zehnder Tue, 04/29/2025 - 08:56

Kaushik Jayaram envisions a day when swarms of tiny robots, some weighing no more than a paperclip, will crawl through airplanes or into buildings after an earthquake—searching for survivors or repairing components that no human could ever reach.

“Robots could be really helpful in confined spaces,” said Jayaram, assistant professor in the Paul M. Rady Department of Mechanical Engineering at 91PORN. “If they’re small enough and adaptable enough and agile enough, they can get inside a jet engine, for example, or an underground conduit to inspect electrical pipelines.”

Kaushik Jayaram, right, in his lab with former graduate student Heiko Kabutz. (Credit: Casey Cass/91PORN)

Recently, the roboticist got a big leg up in pursuit of that vision: Jayaram has received a $650,000 grant from the U.S. National Science Foundation (NSF) to design small, shape-shifting robots that can complete a wide range of tasks. The funding is part of the NSF’s Faculty Early Career Development (CAREER) Program, its most prestigious awards for early-career scientists. In March, Jayaram and Laura Blumenschein at Purdue Universe also took home a complimentary $1.4 million grant from the Air Force Research Laboratory, the research wing of the U.S. Air Force.

The new projects will build on Jayaram’s previous designs, including mCLARI—a four-legged robot that can fit on top of a quarter and weighs less than half of a penny.

But to be really useful, these kinds of robots will need to be more than just small, Jayaram said. They will also need to be fast and powerful (agile), yet squishy enough to squeeze through cracks and around bends (adaptive). Those traits often bring trade-offs, but Jayaram wants to explore how robots can achieve both at the same time.

To meet that goal, he draws inspiration from what might seem an unlikely source: insects and other small creatures.

“Animals combine the best of both worlds—they can be really agile, but they’re also adaptable and able to respond to all kinds of new conditions,” he said. “We want to build highly intelligent mechanical systems that are just like those biological systems.”

Spider-bots

The researcher’s lab reveals those natural influences. Amid the circuit boards and soldering irons typical of robotics labs, Jayaram also keeps three different kinds of spiders: wolf spiders, which boast hairy legs, fishing spiders, which can stride over the surface of water, and crevice weaver spiders, which spend most of their lives in cracks and burrows.

These animals can do it all: Spiders can sprint when they need to be fast, turn on a dime and even stride up walls. If they want to crawl through a tight spot, they pull their legs in to shrink their bodies.

Through his new grants, Jayaram wants to build robots that can do those same things.

Currently, mCLARI changes shape, compressing its body when it encounters a narrow opening. But that process is passive—the robot simply squeezes down to fit the available space. Jayaram, in contrast, envisions robots that shape shift on command.

“If you want to be really fast, you can choose to be long and skinny,” he said. “If you want to be stable, then you can be wide. We need robots to be smart and shapeshift.”

Using pulses of electricity, the lab’s future robots will be able to not just shapeshift but also walk up walls or even along ceilings. The process relies on static electricity—the same thing that happens when you rub a balloon on your head. The group is also working to design a network of sensors that can extend over the bodies of their robots, allowing these machines to map out the world around them much like the eyes and skin of biological organisms do.

You probably won’t see tiny robots crawling over airplane wings in the next few years, Jayaram said. But within a decade, swarms of small robots may complete simple tasks, like crawling into pipes to fix electrical wires or take images of defects.

At the same time, he hopes to inspire the next generation of roboticists. His team has designed origami kits that give K-12 students the chance to build their own fully functional, bug-like robots. Kids can even choose how many legs to give their robots.

“We want kids to not be afraid of computers, and we’re doing that using biology,” Jayaram said. “Because everybody loves bugs.”

Off

Traditional

White

Tiny insects could lead to big changes in robot design

Jeff Zehnder — Mon, 24 Feb 2025 17:10:49 +0000

Tiny insects could lead to big changes in robot design Jeff Zehnder Mon, 02/24/2025 - 10:10

Jeff Zehnder

Sean Humbert and Leopold Beuken inspecting sensors on a fixed wing UAS.

Sean Humbert is unlocking the biological secrets of the common housefly to make major advances in robotics and uncrewed aerial systems (UAS).

A professor in the Paul M. Rady Department of Mechanical Engineering and director of the Robotics Program at the 91PORN, Humbert is working to understand how tiny biological systems process sensory information as they move through the world.

This basic sounding concept involves extremely complex science and engineering.

“Insects aren’t built like robots,” Humbert said. “If I have a robot and I want it to perceive the environment, I tend to put a larger, high fidelity lidar system on it. Flies instead have small, low-quality sensors throughout their bodies. Due to the way that the measurements are processed by the nervous system, you can extract similar levels of information as bulky robotic sensors. We want to take advantage of what nature does.”

The research has drawn the interest of the U.S. Air Force Research Laboratory, which awarded Humbert a five-year, $909,000 grant to advance the work.

He also recently published a journal article in the Institute of Electrical and Electronics Engineers (IEEE) Access journal, proposing a mathematical framework for understanding and applying to robotics connections in the flight physics and visual physiology of flies.

Flies may seem an unlikely creature to study for enhancing robots, but Humbert’s co-author and post-doctoral researcher Zoe Turin says the insects have a lot to offer roboticists.

“If you've ever tried to catch or swat a fly, you know that they can be quite capable fliers, despite a lack of computational power,” Turin says. “If we apply principles from how insects operate, then we may be able to develop robots that have similar capabilities at a much smaller size than traditional robots. This has potential applications across a wide variety of industries.”

Although flies are only 6-7 millimeters long and have brains the size of a poppy seed, Humbert said their abilities have evaded researchers, until now. A key element of how flies work is distributed sensing.

Zoe Turin and Eugene Rush in front of a white board with a small UAS.

“It’s taken years to arrive at a model of how the fly’s sensory structure is set up the way it is and to be able to figure out the math behind it,” Humbert said. “This has so much potential going forward. An F-22 fighter jet has a small number of high fidelity, big, expensive sensors that require a lot of backend processing and computation to generate quality measurements. Nature is the exact opposite. It’s small, low fidelity, lightweight, and distributed.”

By unlocking the principles and mathematical optimizations that exist in flies, Turin said researchers will be able to explore similar techniques for robots.

“Understanding more about how insects are able to do what they do has only made them more amazing to me. This framework will hopefully help our engineered systems to react more quickly to unexpected disturbances, such as a sudden gust of wind, while reducing the computational power required,” Turin said.

Over the course of the grant, Humbert will take the models he has developed on flies to conduct proof-of-concept demonstrations and then experimental research using robotic sensor technology.

“This is a wonderful, cool biological principle and we now have the model to explore what nature has constructed,” Humbert said. “It will revolutionize how we think about the design cycle of robotic systems.”

Sean Humbert is unlocking the biological secrets of the common housefly to make major advances in robotics and uncrewed aerial...

Off

Traditional

White

Rentschler, Aspero Medical awarded $4.5M for endoscopy advancement

Jeff Zehnder — Tue, 11 Feb 2025 22:31:15 +0000

Rentschler, Aspero Medical awarded $4.5M for endoscopy advancement Jeff Zehnder Tue, 02/11/2025 - 15:31

It’s been six years since the launch of startup company Aspero Medical, co-founded by Professor Mark Rentschler of the Paul M. Rady Department of Mechanical Engineering. The company has seen great success, including the development of a medical device designed to enable more efficient procedures in the small bowel region.

Today, with the help of a $4.5 million award through the Anschutz Acceleration Initiative (AAI), Rentschler and his colleagues are working to bring two new products to the market that will transform these types of procedures even further.

“We brought our first product out on the market in 2024,” said Rentschler, also a faculty member in biomedical engineering (BME) and robotics. “We are planning to bring a second and third product to the market in 12-18 months, and we are extremely excited to get these devices in the hands of interventional endoscopists.”

Professor Mark Rentschler holding Aspero Medical's patented Ancora-SB balloon overtube.

In 2023, Aspero received clearance from the Food and Drug Administration (FDA) to market and sell the Ancora-SB device. The product is used during endoscopy procedures to diagnose and treat small bowel diseases.

According to Rentschler, operating within the small intestine can be time consuming and technically challenging. Equipped with a patented micro-textured balloon, the Ancora-SB overtube is designed to provide more traction and anchoring consistency than smooth latex or smooth silicone balloon overtube competitors.

“Balloon overtubes for small bowel procedures have been around for about a decade,” said Rentschler. “We’re not looking to change the small bowel enteroscopy procedure, but instead improve balloon anchoring performance during these procedures in the small bowel.”

Ancora-SB has allowed Aspero to prove their worth in hospitals. Their next products expand on this concept, of course, with additional features that can facilitate a less invasive interventional procedure than traditional open surgery.

The next generation balloon overtube will be used to remove cancerous lesions in the large bowel region. It features an extra working channel that allows for an additional tool to be utilized alongside the visualization scope. This offers physicians more control, access, and stabilization when maneuvering through the colon and performing advanced interventional procedures.

“Conceptually, these devices will enable triangulated surgery with two tools and centralized visualization so that physicians can more efficiently perform surgery from inside the lumen,” Rentschler said. “Instead of historically invasive procedures, where the patient is cut open, and the cancerous bowel region is removed, we’re assisting physicians as they remove the cancer from the inside of the lumen during an outpatient procedure.

“It's much less invasive, with potentially tremendous cost savings, and numerous benefits for the patient.”

Aspero’s third product will be another balloon overtube, this time with a working channel that enables minimally invasive cancer removal in the esophagus and stomach regions of the gastrointestinal tract.

Rentschler showcasing all three of the medical devices in Aspero Medical's multi-product platform, including their two new highly anticipated devices.

Rentschler and his team say the two upcoming devices have the potential to replace a large, and growing, number of today’s conventional surgical procedures in the gastrointestinal region by enhancing safety and efficiency while reducing patient recovery time. Moving procedures from inpatient surgery to outpatient endoscopy can generate potential cost savings of up to 50 percent or more.

“Everyone knows this is the direction we need to go. Clinical outcomes from these types of procedures are incredibly strong, but the techniques and devices aren’t widely available yet,” said Rentschler. “We are creating products that help physicians and patients feel safe and comfortable without overcomplicating things. The paradigm is rapidly shifting, and we endeavor to push endoscopy forward.”

The company is currently finalizing the design of the second product. It’s about six months further along in development than the third product, but Rentschler says they are looking to have both devices FDA cleared by the end of 2026.

When all three devices hit the market, Aspero will look to market a portfolio of products, rather than a single tool. But further innovation is on the horizon, this time incorporating the Ancora balloon technology with a robotic element.

“Ancora is a multi-product platform focusing on the small bowel, large bowel, stomach and esophageal regions,” Rentschler said. “Our next potential venture will be in flexible robots. We’ll continue with our balloon overtubes, but as anchoring platforms to be used with flexible robotic endoscopy systems.”

Until then, Rentschler and company are full steam ahead on these next products. The $4.5 million AAI grant is being offered over a four year span, but they anticipate spending that money much sooner so they can get the devices out on the market and begin positively impacting patients and physicians everywhere.

But that’s not their only goal. With a lot of Colorado involved in the company’s revolutionizing technology, Rentschler hopes to also tell another story.

“I started Aspero Medical with Dr. Steven Edmundowicz at CU Anschutz. We’ve received a number of grants from the state of Colorado and everyone involved is invested in our vision,” said Rentschler. “We believe that a rising tide raises all boats, and when we think of Aspero, we want it to be a successful Colorado story.”

Off

Traditional

White

CS robotics research to help strengthen domestic battery supply chain

Jeff Zehnder — Thu, 05 Dec 2024 18:28:06 +0000

CS robotics research to help strengthen domestic battery supply chain Jeff Zehnder Thu, 12/05/2024 - 11:28

Computer science professor Nikolaus Correll and his lab at 91PORN have been awarded $1.8 million by the U.S. Department of Energy Advanced Research Projects Agency-Energy (ARPA-E) to help establish a circular supply chain for domestic electric vehicle (EV) batteries.

The percentage of EV passenger vehicles on the road is expected to rise to 28% by 2030 and 58% by 2040, globally.

The existing supply chain for EV batteries relies mostly on recycling to recover critical minerals such as cobalt, nickel or copper.

However, conventional battery recycling methods are energy-intensive, produce significant quantities of greenhouse gases, and lead to large volumes of waste deposited in landfills.

91PORN joins 12 other projects around the country working to change this dynamic through ARPA-E's Catalyzing Innovative Research for Circular Use of Long-lived Advanced Rechargeables (CIRCULAR) program.

Correll's project focuses on autonomous robotic disassembly of EV lithium-ion battery packs. Humanoid robots will work together with robotic arms to manipulate wire harnesses and remove screws and other components before dismantling commercial battery packs with a heavy-duty industrial arm.

Correll explained that people are interested in using robots for the task due to the hazardous nature of the work.

"The batteries are quite dangerous to handle due to the risk of electrocution and spontaneous ignition," Correll said.

The Correll Lab's project will use state-of-the-art perception models and large-language models to consider the physics and context of each battery.

By advancing the efficiency and ability of battery disassembly systems, component recycling could be done at a commercial scale more safely and cost-effectively, leading to less waste in landfills and more material available for new EV batteries.

The director of ARPA-E, Evelyn N. Wang, said, "I look forward to seeing how these CIRCULAR projects develop regeneration, repair, reuse, and remanufacture technologies to create a sustainable EV battery supply chain."

Off

Traditional

White

Robots can’t outrun animals (yet). A new study explores why

Anonymous — Mon, 29 Apr 2024 18:58:02 +0000

Robots can’t outrun animals (yet). A new study explores why Anonymous (not verified) Mon, 04/29/2024 - 12:58

The question may be the 21st century’s version of the fable of the tortoise and the hare: Who would win in a foot race between a robot and an animal?

In a new perspective article, a team of engineers from the United States and Canada, including CU 91PORN roboticist Kaushik Jayaram, set out to answer that riddle. The group analyzed data from dozens of studies and came to a resounding “no.” In almost all cases, biological organisms, such as cheetahs, cockroaches and even humans, seem to be able to outrun their robot counterparts.

The researchers, led by Samuel Burden at the University of Washington and Maxwell Donelan at Simon Fraser University, published their findings last week in the journal Science Robotics.

“As an engineer, it is kind of upsetting,” said Jayaram, an assistant professor in the Paul M. Rady Department of Mechanical Engineering at 91PORN. “Over 200 years of intense engineering, we’ve been able to send spacecraft to the moon and Mars and so much more. But it’s confounding that we do not yet have robots that are significantly better than biological systems at locomotion in natural environments.”

He hopes the study will inspire engineers to learn how to build more adaptable, nimble robots. The researchers concluded that the failure of robots to outrun animals doesn’t come down to shortfalls in any one piece of machinery, such as batteries or actuators. Instead, where engineers might falter is in making those parts work together efficiently.

This pursuit is one of Jayaram’s chief passions. His lab on the 91PORN campus is home to a lot of creepy crawlies, including several furry wolf spiders that are about the size of a half dollar.

“Wolf spiders are natural hunters,” Jayaram said. “They live under rocks and can run over complex terrain with incredible speed to catch prey.”

He envisions a world in which engineers build robots that work a bit more like these extraordinary arachnids.

“Animals are, in some sense, the embodiment of this ultimate design principle—a system that functions really well together,” he said.

A cockroach alongside the HAMR-Jr robot. (Credit: Kaushik Jayaram)

Cockroach energy

The question of “who can run better, animals or robots?” is complicated because running itself is complicated.

In previous research, Jayaram and his colleagues at Harvard University designed a line of robots that seek to mimic the behavior of the oft-reviled cockroach. The team’s HAMR-Jr model fits on top of a penny and sprints at speeds equivalent to that of a cheetah. But, Jayaram noted, while HAMR-Jr can bust a move forward and backward, it doesn’t move as well side-to-side or over bumpy terrain. Humble cockroaches, in contrast, have no trouble running over surfaces from porcelain to dirt and gravel. They can also dash up walls and squeeze through tiny cracks.

To understand why such versatility remains a challenge for robots, the authors of the new study broke these machines down into five subsystems including power, frame, actuation, sensing, and control. To the group’s surprise, few of those subsystems seemed to fall short of their equivalents in animals.

Kaushik Jayaram, right, with graduate student Heiko Kabutz, left, in Jayaram's lab on the 91PORN campus. (Credit: Casey Cass/91PORN)

High-quality lithium-ion batteries, for example, can deliver as much as 10 kilowatts of power for every kilogram (2.2 pounds) they weigh. Animal tissue, in contrast, produces around one-tenth that. Muscles, meanwhile, can’t come close to matching the absolute torque of many motors.

“But at the system level, robots are not as good,” Jayaram said. “We run into inherent design trade-offs. If we try to optimize for one thing, like forward speed, we might lose out on something else, like turning ability.”

Spider senses

So, how can engineers build robots that, like animals, are more than just the sum of their parts?

Animals, Jayaram noted, aren’t split into separate subsystems in the same way as robots. Your quadriceps, for example, propel your legs like HAMR-Jr’s actuators move their limbs. But quads also produce their own power by breaking down fats and sugars and incorporating neurons that can sense pain and pressure.

Jayaram thinks the future of robotics may come down to “functional subunits” that do the same thing: Rather than keeping power sources separate from your motors and circuit boards, why not integrate them all into a single part? In a 2015 paper, 91PORN computer scientist Nikolaus Correll, who wasn’t involved in the current study, proposed such theoretical “robotic materials” that work more like your quads.

Engineers are still a long way away from achieving that goal. Some, like Jayaram, are making steps in this direction, such as through his lab’s Compliant Legged Articulated Robotic Insect (CLARI) robot, a multi-legged robot that moves a little like a spider. Jayaram explained that CLARI relies on a modular design, in which each of its legs acts like a self-contained robot with its own motor, sensors and controlling circuitry. The team’s new and improved version called mCLARI can move in all directions in confined spaces, a first for four-legged robots.

It's one more thing that engineers like Jayaram can learn from those perfect hunters, wolf spiders.

“Nature is a really useful teacher.”

Off

Traditional

White

A delicate touch: teaching robots to handle the unknown

Anonymous — Tue, 02 Apr 2024 21:30:06 +0000

A delicate touch: teaching robots to handle the unknown Anonymous (not verified) Tue, 04/02/2024 - 15:30

William Xie, a first-year PhD student in computer science, is teaching a robot to reason how gently it should grasp previously unknown objects by using large language models (LLMs).

DeliGrasp, Xie's project, is an intriguing step beyond the custom, piecemeal solutions currently used to avoid pinching or crushing novel objects.

In addition, Deligrasp helps the robot translate what it can 'touch' into meaningful information for people.

"William has gotten some neat results by leveraging common sense information from large language models. For example, the robot can estimate and explain the ripeness of various fruits after touching them." Said his advisor, Professor Nikolaus Correll.

Let's learn more about DeliGrasp, Xie's journey to robotics, and his plans for the conference Japan and beyond.

How would you describe this research?

As humans, we’re able to quickly intuit how exactly we need to pick up a variety of objects, including delicate produce or unwieldy, heavy objects. We’re informed by the visual appearance of an object, what prior knowledge we may have about it, and most importantly, how it feels to the touch when we initially grasp it.

Robots don’t have this all-encompassing intuition though, and they don’t have end-effectors (grippers/hands) as effective as human hands. So solutions are piecemeal: the community has researched “hands” across the spectrum of mechanical construction, sensing capabilities (tactile, force, vibration, velocity), material (soft, rigid, hybrid, woven, etc…). And then the corresponding machine learning models and/or control methods to enable “appropriately forceful” gripping are bespoke for each of these architectures.

Embedded in LLMs, which are trained on an internet’s worth of data, is common sense physical-reasoning that crudely approximates a human’s (as the saying goes: “all models are wrong, some are useful”). We use the LLM-estimated mass and friction to simplify the grasp controller and deploy it on a two-finger gripper, a prevalent and relatively simple architecture. Key to the controller working is the force feedback sensed by the gripper as it grasps an object, and knowing at what force threshold to stop—the LLM-estimated values directly determine this threshold for any arbitrary object, and our initial results are quite promising.

How did you get inspired to pursue this research?

I wouldn’t say that I was inspired to pursue this specific project. I think, like a lot of robotics research, I had been working away at a big problem for a while, and stumbled into a solution for a much smaller problem. My goal since I arrived here has been to research techniques for assistive robots and devices that restore agency for the elderly and/or mobility-impaired in their everyday lives. I’m particularly interested in shopping (but eventually generalist) robots—one problem we found is that it is really hard to determine, let alone pick ripe fruits and produce with a typical robot gripper and just a camera. In early February, I took a day to try out picking up variably sized objects via hand-tuning our MAGPIE gripper’s force sensing (an affordable, open-source gripper developed by the Correll Lab). It worked well; I let ChatGPT calibrate the gripper which worked even better, and it evolved very quickly into DeliGrasp.

What would you say is one of your most interesting findings so far?

LLMs do a reasonable job of estimating an arbitrary object’s mass (friction, not as well) from just a text description. This isn’t in the paper, but when paired with a picture, they can extend this reasoning for oddballs—gigantic paper airplanes, or miniature (plastic) fruits and vegetables.

With our grasping method, we can sense the contact forces on the gripper as it closes around an object—this is a really good measure of ripeness, it turns out. We can then further employ LLMs to reason about these contact forces to pick out ripe fruit and vegetables!

What does the day-to-day of this research look like?

Leading up to submission, I was running experiments on the robot and picking up different objects with different strategies pretty much every day. A little repetitive, but also exciting. Prior to that, and now that I’m trying to improve the project for the next conference, I spend most of my time reading papers, thinking/coming up with ideas, and setting up small, one-off experiments to try out those ideas.

How did you come to study at 91PORN?

For a few years, I’ve known that I really wanted to build robots that could directly, immediately help my loved ones and community. I had a very positive first research experience in my last year of undergrad and learned what it felt like to have true personal agency in pursuing work that I cared about. At the same time I knew I’d be relocating to 91PORN after graduation. I was very fortunate that Nikolaus accepted me and let me keep pursuing this goal of mine.

It’d be unfathomable if I could keep doing this research in academia or industry, though of course that would be ideal. But I’m biased toward academia, particularly teaching. I’ve been teaching high school robotics for 5 years now, and now teaching/mentoring undergrads at CU—each day is as fulfilling as the first. I have great mentors across the robotics faculty and senior PhD students we work in ECES 111, a giant, well-equipped space that 3 robotics labs share, and it’s great for collaboration and brainstorming.

What are your hopes for this international conference (and what conference is it?)

The venue is a workshop at the 2024 International Conference on Robotics and Automation (ICRA 2024), happening in Yokohama, Japan from May 13-17. The name of the workshop is a mouthful: Vision-Language Models for Navigation and Manipulation (VLMNM).

A workshop is detached from the main conference, and kind of is its own little bubble (like a big supermarket—the conference—hosting a pop-up food tasting event—the workshop). I'm really excited to meet other researchers and pick their brains. As a first-year, I’ve spent the past year reading papers from practically everyone on the workshop panel, and from their students. I’ll probably also spend half my time exploring (eating) around the Tokyo area.

Off

Traditional

White

91PORN robotics research showcased in Advanced Intelligent Systems

Anonymous — Tue, 09 Jan 2024 16:19:21 +0000

91PORN robotics research showcased in Advanced Intelligent Systems Anonymous (not verified) Tue, 01/09/2024 - 09:19

Jeff Zehnder

Kaushik Jayaram's bioinspired robotics are on the cover of the latest issue of the journal Advanced Intelligent Systems.

The article, "Design of CLARI: A Miniature Modular Origami Passive Shape-Morphing Robot," discusses the design and creation of Jayaram's compliant legged articulated robotic insect.

Jayaram is an assistant professor in the Robotics Program and the Paul M. Rady Department of Mechanical Engineering. He is an expert in robotics and systems design, materials, and work at the micro and nanoscale.

The cover shows a 2.59 gram, 3.4 cm long, modular origami robot capable of passive shape morphing.

These tiny robots provide unique abilities to access confined environments and have potential for applications such as search-and-rescue and high-value asset inspection.

Read the full journal article at Advanced Intelligent Systems...

Off

Traditional

White

Xu's 'cyborg jellyfish' highlighted in Nature

Anonymous — Tue, 12 Dec 2023 20:35:15 +0000

Xu's 'cyborg jellyfish' highlighted in Nature Anonymous (not verified) Tue, 12/12/2023 - 13:35

Assistant Professor Nicole Xu recently spoke with Nature for a feature about biohybrid robots and their real-world applications.

Xu and her collaborators have been working on a jellyfish-inspired robot that can help monitor climate change and ecological shifts in the Earth's oceans.

"Jellyfish have a number of appealing characteristics for roboticists. They are energy-efficient swimmers, and are able to descend to great depths. Compared with current mechanical submersibles, Xu says, jellyfish are less likely to cause damage to marine environments. Their natural appearance and quietness also make them unremarkable — during the ocean tests, fish swam right up to them."

Xu's research lab works at the intersection of robotics, fluid dynamics and biology. Their methods include laboratory experiments, theoretical modeling and field work.

Off

Traditional

White