#magnetic grippers

[mpcomagnetics]

HIJAB MAGNETS VS HIJAB MAGNETIC PINS

HIJAB MAGNETS VS HIJAB MAGNETIC PINS The fashion world is constantly evolving, and one of the notable innovations is Hijab Magnets. These small, powerful accessories are redefining how hijabs are worn, merging convenience with style. This detailed exploration delves into the benefits, safety, varieties, and transformative impact of Hijab Magnets on the traditional hijab, offering insights into…

[teaboot]

Grandpa-Related Post

Okay so if you saw this post then you know that my brother and I have located our AWOL grandpa

Now, I kniw he's always had problems with his hands, but ten years later they are borderline useless and it's definitely a problem

(See diagram below)

now, his nails are all very thick and long and that isn't helping, but that's something we can fix. What I'm more concerned about is the swelling and permanent curl in his index fingers, and the locked joints in his thumbs.

The three furthest fingers in each hand are constantly sort of fisted as far as I can tell, and the wrists themselves are fused straight after several surgeries and no longer bend.

What he can do now on his own is limited to thumb-index pinching movements, and kind of hooking things with his curled fingers.

He can fold paper, hold a sandwich, and pull stuff out of his (unzipped, open, never closed) fanny pack.

Right now, he isn't able to move fast despite being (obviously) highly independant, and essentially spends all day outside his care home wandering and loitering in the sketchy part of town with cash visibly sticking out his bag.

So, I'm looking to see if anyone has any ideas on accessibility aids or bags or grippers or anything he can use to help him do things on his own.

I might be able to modify a bag to close with magnets or velcro instead of a zipper, and am already looking for a dual-handled cup that doesn't look like it's meant for toddlers.

He has slipped a wee bit in the memory area but is otherwise very present- though he has stated that he's bored, lonely, and mentally understimulated.

His chief complaint at the moment is that the food at his care home tastes like shit, and "everyone else has dementia and can't carry a decent conversation".

Also he can't hold his phone anymore and his service was cut, but I have a few ideas there. Still open to suggestions, though.

Input welcome!!

[arctic-hands]

PEN GEEKS I NEED HELP

What's a good pen analogous to the the Uni-Ball Signo 207? It's my favorite daily pen: quick drying, easy and smooth gel ink depositing for aching wrist and finger joints, a clicky top (tho I'll take a twisty bottom, just nothing with a cap I can lose on the go), waterproof and archival, perfect slimness for my smallish hands, AND it comes in a .38 Ultra Micro nib which is essential because I write in very small cursive. But it comes with a bulky silicone gripper which doesn't fit in the elastic loop of most planners--including my current bujo shell--at least not without struggle getting it in or out. Not great if I'm taking important notes in front of someone in a hurry (such as doctors)

At some point I am going to replace the current filofax (knockoff) shell and it's inflexible magnetic closure for another filofax knockoff that has an elastic closure around it, so a pen with a clip could just go around the elastic, but the one I want is more expensive than a new pen. Plus, now I get to experiment with new pens!

[duketibbitswaifu]

duke tibbits feet. like if agree

🌸 sorry babes, ✨ not into feet. 🍰 put those grippers AWAY~111 🌸

💗 What I do like though is his musk. 🎀 Fuckkkk if I'm metal, 💕 then Duke pits are magnetic. 🧲 🥰 Especially when he gets home from work sweatin' all day. ✨ And when he unbuckles his jeans 🍓 and lets me sniff the bulge in his soaked tighty whities? 🩲 Fuck man, I'm out. 🤍 Nothing better than the smell of sweat, cigarettes, and precum. 🌺 He really knows how to rile me up. 💗

art credit @yeenthebin

[superstupendoussims]

Clyde was tired of constantly having to clean up after herself and her messy children. she wished there was an easier way to keep their house clean. That's when she had an idea - she would build a robot that would clean up messes in the house.Clyde was a skilled inventor and had a workshop on her front porch where she spent most of her free time tinkering with various gadgets and machines. she gathered all the necessary materials and got to work on her new project.

The first thing she did was design the robot's body.she wanted it to be compact and lightweight so it could easily maneuver around the house. she also wanted it to have a sleek and modern look, so she used shiny metal and bright colors for the exterior.

Next, she worked on adding a hovering mechanism. she used powerful magnets and a propulsion system that would allow the robot to float and move around the house without any wheels or legs. This would also make it easier for the robot to reach high and narrow spaces.For the cleaning function, Clyde installed a vacuum system and a set of brushes that could sweep and mop the floors. she also added a detachable arm with a hand-like gripper that could pick up objects and put them in the trash.

Once the robot's body and functions were complete, Clyde programmed it with a set of instructions that would allow it to clean efficiently. she made sure it could detect messes and navigate around furniture and obstacles.she also added a voice recognition system so she could give commands to the robot. After weeks of hard work, the hovering robot was finally complete. Clyde couldn't wait to test it out.she turned it on and watched as it floated off the ground and started cleaning the room. The robot moved swiftly and efficiently, picking up dirt and debris as it went. Clyde was amazed at how well it worked.

[atplblog]

Price: [price_with_discount] (as of [price_update_date] - Details) [ad_1] Tummy Twister waist trimmer ab exercise Tummy Twister: The ab exerciser targets specific muscles of your body to build your fitness. Rowing movements of this exerciser trims and tones your abdominal muscles and strengthens your arms, legs, hips and thighs. It burns extra calories, melts away your abdominal fat around your waist to give you a flat and firm stomach. The exerciser also functions as a hand gripper. Grip the inner handle and pull. It exercises your arms, wrists, fingers, forearms simultaneously Portable Fitness Equipment: The ab exerciser is portable fitness equipment that can be used in your home or gym comfortably. It comes fully assembled for immediate use. The steel springs can be removed easily and put back on the expander. It is beneficial for beginners and advanced users. Make your body muscles more strong and get an amazing physique with this tummy twister from the house of GJSHOP This tummy twister helps you to burn your calories and make you help for developing a flat stomach It also works simultaneously on your abs and core muscles by melting your fatThis twister can be used to tighten your legs, bellies and back.The plast material of this twister firmer your forearms, sleeves and fingers and provides a good Foot grip. For an amazing body shape also Magnetic acupressure Waist Twister Exercise boosts your metabolism. Best moves for a slim waist, circulation and reduces muscular aches and pains. Precision ball-bearing ride for smooth, fluid motion, High-impact, rotating platform provides a vigorous form of aerobic exercise. Dimension Tummy Trimmer :-Length - 14.5 Inches Material - Stainless Steel Tummy Twister :-Measures:9.8x1.1 inch.Weight:0.5KG.Weight Capacity:100 kg Color :- Multicolour As per Availability Abdomen, Arm & Chest Exercise]: Both standing and sitting using are ok. Different using method for different exercise. Ideal for toning & strengthening stomach, waist and legs, arms, hips, thighs and works on the tummy at the same time Lose Fat, Shape your Body]: This resistance training tool workouts your arms, tummy, shoulders, legs and butt. It is not only way to lose your fat by exercising with our tummy trimmer but also a way to shape your body Upgrade, MORE Durable]: Made of upgrade high quality plastic and spring, much stronger than the old one, durable and difficult to crack, best choice for building up your muscles Package Includes, 3 PC SET]: 1 Tummy Trimmer & 1 Tummy Twister & 1 Toning Tube [ad_2]

[moonlightchn]

¡𝕸𝖊𝖗𝖗𝖞 𝕮𝖍𝖗𝖎𝖘𝖙𝖒𝖆𝖘!

                      🌓 𝖂𝖊𝖗𝖊𝖜𝖔𝖑𝖋 𝕮𝖍𝖆𝖓𝖓𝖎𝖊

For: @clubwnderland Jongin 🐺

The wolf leaned quietly against the doorframe of the living room, hands in the pockets of the hoodie he stole, eyes fixed on the way his boyfriend's hands pushed back his hair as he laughed at something his mom said.

Perfect. It was just... Perfect. Magical. Heaven. How many more words are there to describe the feeling that overcomes your heart when the person you love the most is loved by and loves those you love? Uh... Messy thought.

Smiling as his mom now noticed him and motioned the boy over, Channie made his way around the couch to plop down next to his boyfriend, lips kissing his softly before he joined the conversation.

As always, presents started with the youngest, the siblings opening up their own and showing off before Channie's turn came around.

A huge smile adorned the wolf's lips as he pulled out the frog oodie, eyes lightening up as he brought the fabric to his face, cheek rubbing on the softness.

The little knitted beanie right after made him giggle, of course it was Alice. He wondered if she'd like her gift too! She promised she wouldn't open it before Christmas!

The mugs were the last to be opened, their own silly thing as he chuckled and showed them off to his parents with a soft screech before turning to kiss Jongin yet again. And again. And again.

He always had the best gifts.

Jongin came right after, now Channie wiggling nervously in place as his siblings handed over the packages with Jongin's name.

The first he opened was the bag, inside the pair of fuzzy socks and magnet socks.

"Listen, listen, listen! I know you like having your grippers out, but! We can hold hands with socks now! And! Puppy socks! " In Channie's mind, they kinda just... Made sense.

Next up was the airpods case, small dolphin added, and cute keychain to adorn it.

"Isn't it so cute! I- I know it's not for your earpods, silly. That's why I bought you some too." The wolf rolled his eyes as if it was obvious, hands reaching to steal the case from his boyfriend as he opened it up to show the brand new devices inside.

Last, but not least, the bigger box. Channie wiggled nervously, his eyes flickering from the paper to him as Jongin ripped it slowly to reveal the navy blue packaging, gold letters reading the brand right on top, red bow holding it all together.

"Open it! Go! Go! " Channie bounced in place, not really having seen the finished product either as it arrived directly at the house. His hands gripped his boyfriend's arm as he brought out the custom-made cardigan, his initials on the side.

Channie screeched, excited at how pretty and soft the piece was.

"Merry Christmas, Wolfman!!! "

[hsmagnet]

Application of Channel Magnet Assembly

Application of Channel Magnet Assembly Channel magnets can be used as Rectangular Magnet Base, Magnetic Door Latches, Vehicle License Plate Holders, Magnetic Sign and Banner Holders, Magnetic Ceiling Hooks, etc. Use your creative mind and there is no limitation about how these channel magnets can be used. Here are some examples: Those channel magnets with grippers are used in offices, stores to…

[mpcomagnetics]

What Are Magnetic Hijab Pins?

Magnetic hijab pins are a convenient and stylish way to secure your hijab without damaging the fabric. Unlike traditional pins, they don’t leave holes or snags, making them a perfect choice for delicate materials. This guide will walk you through the steps to use them effectively, ensuring your hijab stays in place all day while maintaining a polished look. I remember the first time I switched to…

[12-548]

Robot Gripping System Market Growth and Upcoming Trends 2024-2032

Global "Robot Gripping System Market" report has witnessed |Steady and Robust Growth 2024-2032| in recent years and is anticipated to maintain this optimistic progression until 2032. One notable trend within the Automotive Tool Holder market is the growing preference for sustainable and eco-friendly products. Another significant trend in the Automotive Tool Holder market is the escalating integration of technology to enhance product quality and efficiency.

Who is the largest manufacturers of Robot Gripping System Market worldwide?

FIPA (Germany)

SMC (Japan)

Bastian Solutions (U.S.)

Schmalz (Germany)

Destaco (U.S.)

EMI (U.S.)

SAS Automation (U.S.)

Soft Robotics (U.S.)

Robotiq (Canada)

Schunk (Germany)

Applied Robotics (U.S.)

Zimmer (Germany)

Festo (Germany)

IAI (Japan)

Grabit (U.S.)

RAD (U.S.)

Request a Free Sample Copy @ - https://www.globalmarketstatistics.com/enquiry/request-sample-pdf/10908

Short Description About Robot Gripping System Market:

The Global Robot Gripping System market is anticipated to rise at a considerable rate during the forecast period, between 2024 and 2032. In 2023, the market is growing at a steady rate and with the rising adoption of strategies by key players, the market is expected to rise over the projected horizon.

North America, especially The United States, will still play an important role which cannot be ignored. Any changes from United States might affect the development trend of Co-Living. The market in North America is expected to grow considerably during the forecast period. The high adoption of advanced technology and the presence of large players in this region are likely to create ample growth opportunities for the market.

Europe also play important roles in global market, with a magnificent growth in CAGR During the Forecast period 2024-2032.

Co-Living Market size is projected to reach Multimillion USD by 2032, In comparison to 2024, at unexpected CAGR during 2024-2032.

Despite the presence of intense competition, due to the global recovery trend is clear, investors are still optimistic about this area, and it will still be more new investments entering the field in the future.

This report focuses on the Co-Living in global market, especially in North America, Europe and Asia-Pacific, South America, Middle East and Africa. This report categorizes the market based on manufacturers, regions, type and application.

The report focuses on the Co-Living market size, segment size (mainly covering product type, application, and geography), competitor landscape, recent status, and development trends. Furthermore, the report provides detailed cost analysis, supply chain.

Technological innovation and advancement will further optimize the performance of the product, making it more widely used in downstream applications. Moreover, Consumer behavior analysis and market dynamics (drivers, restraints, opportunities) provides crucial information for knowing the Co-Living market.

What are the types of Robot Gripping System available in the Market?

Based on Product Types the Market is categorized into Below types that held the largest Robot Gripping System market share In 2024.

Electric Grippers

Pneumatic Grippers

Vacuum Grippers/Suction Cups

Magnetic Grippers

Request a Free Sample Copy @ - https://www.globalmarketstatistics.com/enquiry/request-sample-pdf/10908

Which regions are leading the Robot Gripping System Market?

North America (United States, Canada and Mexico)

Europe (Germany, UK, France, Italy, Russia and Turkey etc.)

Asia-Pacific (China, Japan, Korea, India, Australia, Indonesia, Thailand, Philippines, Malaysia and Vietnam)

South America (Brazil, Argentina, Columbia etc.)

Middle East and Africa (Saudi Arabia, UAE, Egypt, Nigeria and South Africa)

This Robot Gripping System Market Research/Analysis Report Contains Answers to your following Questions

What are the global trends in the Robot Gripping System market? Would the market witness an increase or decline in the demand in the coming years?

What is the estimated demand for different types of products in Robot Gripping System? What are the upcoming industry applications and trends for Robot Gripping System market?

What Are Projections of Global Robot Gripping System Industry Considering Capacity, Production and Production Value? What Will Be the Estimation of Cost and Profit? What Will Be Market Share, Supply and Consumption? What about Import and Export?

Where will the strategic developments take the industry in the mid to long-term?

What are the factors contributing to the final price of Robot Gripping System? What are the raw materials used for Robot Gripping System manufacturing?

How big is the opportunity for the Robot Gripping System market? How will the increasing adoption of Robot Gripping System for mining impact the growth rate of the overall market?

How much is the global Robot Gripping System market worth? What was the value of the market In 2023?

Who are the major players operating in the Robot Gripping System market? Which companies are the front runners?

Which are the recent industry trends that can be implemented to generate additional revenue streams?

What Should Be Entry Strategies, Countermeasures to Economic Impact, and Marketing Channels for Robot Gripping System Industry?

Request Free Sample PDF @ - https://www.globalmarketstatistics.com/enquiry/request-sample-pdf/10908

About Us:

Global Market Statistics is a unique organization that offers expert analysis and accurate data-based market intelligence, aiding companies of all shapes and sizes to make well-informed decisions. We tailor inventive solutions for our clients, helping them tackle any challenges that are likely to emerge from time to time and affect their businesses.

For More Related Reports Click Here :

Aesthetic Services Market

Robot Gripping System Market

Minimally Invasive Glaucoma Surgery (MIGS) Devices Market

Probiotic Drinks Market

Automation in Automotive Market

Application Container Market

Holographic Anti-counterfeiting Marks Market

Managed Security Services (MSS) Market

Commercial Telematics Market

Safety Lancets Market

WAN Optimization Market

Coworking Spaces Market

Dental Bone Void Filler Market

Autopilot System Market

Cell Culture Media, Sera, and Reagents Market

[imfixedonengineering]

Mechanical Aids for Your Business Environments!

There are many tools that can be used in a business environment in order to improve the efficiency and safety of an organisation. A lot of these tools fall under the category of mechanical aids.

In this blog, I will explain the types of mechanical aids that are available to your business and how they work.

End Effectors

End effectors are tools or devices attached to the end of a robotic arm that interact with objects to perform specific tasks. They can be customized for various applications.

Types:

Grippers: These are used to grasp and hold objects. They can be mechanical, vacuum-based, magnetic, or adhesive.

Welders: Used in robotic welding applications, these end effectors include spot welders, arc welders, and laser welders.

Tools: This category includes screwdrivers, drills, cutters, and other tools that a robot may need to perform specific tasks.

Sensors: Some end effectors include sensors to measure properties like force, torque, or proximity, providing feedback to the robot.

Specialized Devices: Custom end effectors designed for specific tasks, such as painting, dispensing fluids, or handling delicate materials.

Importance of End Effectors

Task Performance: End effectors are essential for a robot to interact with and manipulate its environment. They determine the range of tasks a robot can perform, from simple pick-and-place operations to complex assembly and processing tasks.

Flexibility: Robots can be equipped with different end effectors to handle various tasks, making them versatile and adaptable to different applications within manufacturing, healthcare, logistics, and more.

Efficiency: The use of end effectors in automation enhances efficiency by speeding up processes, reducing errors, and enabling continuous operation without fatigue.

Precision: End effectors can perform tasks with high precision, which is critical in industries like electronics, pharmaceuticals, and automotive manufacturing where exactness is paramount.

Safety: By using robots with appropriate end effectors, dangerous tasks can be automated, reducing the risk of injury to human workers and allowing humans to focus on safer, more strategic activities.

Stairclimbers

These machines are battery powered transporters that can allow the transportation of heavy objects.

The Ninja Stairclimber from ACES is a good example of the maximum capabilities of stairclimbers with two speed settings, single step, and continuous-step modes that allow you to move heavy objects up and down a set of stairs. The step edge brake system guarantees the highest level of safety with an auto detection system which prevents any user’s feet from being entrapped whilst in operation.

Their Zeus model is also reliable, with an extended lifting height which can be operated by one person. This makes lifting and transporting heavy objects safer and easier for all users. The Zeus is ideal for confined spaces, simple to use and combines lifting and transporting functions up to a height of 1600mm as standard.

Conclusion

These two mechanical aids can allow for a more efficient and safer work environment, as heavy objects can be transported without risk of injury or delay.

[sunaleisocial]

A better way to control shape-shifting soft robots

New Post has been published on https://sunalei.org/news/a-better-way-to-control-shape-shifting-soft-robots/

A better way to control shape-shifting soft robots

Imagine a slime-like robot that can seamlessly change its shape to squeeze through narrow spaces, which could be deployed inside the human body to remove an unwanted item.

While such a robot does not yet exist outside a laboratory, researchers are working to develop reconfigurable soft robots for applications in health care, wearable devices, and industrial systems.

But how can one control a squishy robot that doesn’t have joints, limbs, or fingers that can be manipulated, and instead can drastically alter its entire shape at will? MIT researchers are working to answer that question.

They developed a control algorithm that can autonomously learn how to move, stretch, and shape a reconfigurable robot to complete a specific task, even when that task requires the robot to change its morphology multiple times. The team also built a simulator to test control algorithms for deformable soft robots on a series of challenging, shape-changing tasks.

Play video

Their method completed each of the eight tasks they evaluated while outperforming other algorithms. The technique worked especially well on multifaceted tasks. For instance, in one test, the robot had to reduce its height while growing two tiny legs to squeeze through a narrow pipe, and then un-grow those legs and extend its torso to open the pipe’s lid.

While reconfigurable soft robots are still in their infancy, such a technique could someday enable general-purpose robots that can adapt their shapes to accomplish diverse tasks.

“When people think about soft robots, they tend to think about robots that are elastic, but return to their original shape. Our robot is like slime and can actually change its morphology. It is very striking that our method worked so well because we are dealing with something very new,” says Boyuan Chen, an electrical engineering and computer science (EECS) graduate student and co-author of a paper on this approach.

Chen’s co-authors include lead author Suning Huang, an undergraduate student at Tsinghua University in China who completed this work while a visiting student at MIT; Huazhe Xu, an assistant professor at Tsinghua University; and senior author Vincent Sitzmann, an assistant professor of EECS at MIT who leads the Scene Representation Group in the Computer Science and Artificial Intelligence Laboratory. The research will be presented at the International Conference on Learning Representations.

Controlling dynamic motion

Scientists often teach robots to complete tasks using a machine-learning approach known as reinforcement learning, which is a trial-and-error process in which the robot is rewarded for actions that move it closer to a goal.

This can be effective when the robot’s moving parts are consistent and well-defined, like a gripper with three fingers. With a robotic gripper, a reinforcement learning algorithm might move one finger slightly, learning by trial and error whether that motion earns it a reward. Then it would move on to the next finger, and so on.

But shape-shifting robots, which are controlled by magnetic fields, can dynamically squish, bend, or elongate their entire bodies.

The researchers built a simulator to test control algorithms for deformable soft robots on a series of challenging, shape-changing tasks. Here, a reconfigurable robot learns to elongate and curve its soft body to weave around obstacles and reach a target.

Image: Courtesy of the researchers

“Such a robot could have thousands of small pieces of muscle to control, so it is very hard to learn in a traditional way,” says Chen.

To solve this problem, he and his collaborators had to think about it differently. Rather than moving each tiny muscle individually, their reinforcement learning algorithm begins by learning to control groups of adjacent muscles that work together.

Then, after the algorithm has explored the space of possible actions by focusing on groups of muscles, it drills down into finer detail to optimize the policy, or action plan, it has learned. In this way, the control algorithm follows a coarse-to-fine methodology.

“Coarse-to-fine means that when you take a random action, that random action is likely to make a difference. The change in the outcome is likely very significant because you coarsely control several muscles at the same time,” Sitzmann says.

To enable this, the researchers treat a robot’s action space, or how it can move in a certain area, like an image.

Their machine-learning model uses images of the robot’s environment to generate a 2D action space, which includes the robot and the area around it. They simulate robot motion using what is known as the material-point-method, where the action space is covered by points, like image pixels, and overlayed with a grid.

The same way nearby pixels in an image are related (like the pixels that form a tree in a photo), they built their algorithm to understand that nearby action points have stronger correlations. Points around the robot’s “shoulder” will move similarly when it changes shape, while points on the robot’s “leg” will also move similarly, but in a different way than those on the “shoulder.”

In addition, the researchers use the same machine-learning model to look at the environment and predict the actions the robot should take, which makes it more efficient.

Building a simulator

After developing this approach, the researchers needed a way to test it, so they created a simulation environment called DittoGym.

DittoGym features eight tasks that evaluate a reconfigurable robot’s ability to dynamically change shape. In one, the robot must elongate and curve its body so it can weave around obstacles to reach a target point. In another, it must change its shape to mimic letters of the alphabet.

In this simulation, the reconfigurable soft robot, trained using the researchers’ control algorithm, must change its shape to mimic objects, like stars, and the letters M-I-T.

Image: Courtesy of the researchers

“Our task selection in DittoGym follows both generic reinforcement learning benchmark design principles and the specific needs of reconfigurable robots. Each task is designed to represent certain properties that we deem important, such as the capability to navigate through long-horizon explorations, the ability to analyze the environment, and interact with external objects,” Huang says. “We believe they together can give users a comprehensive understanding of the flexibility of reconfigurable robots and the effectiveness of our reinforcement learning scheme.”

Their algorithm outperformed baseline methods and was the only technique suitable for completing multistage tasks that required several shape changes.

“We have a stronger correlation between action points that are closer to each other, and I think that is key to making this work so well,” says Chen.

While it may be many years before shape-shifting robots are deployed in the real world, Chen and his collaborators hope their work inspires other scientists not only to study reconfigurable soft robots but also to think about leveraging 2D action spaces for other complex control problems.

[jcmarchi]

A better way to control shape-shifting soft robots

New Post has been published on https://thedigitalinsider.com/a-better-way-to-control-shape-shifting-soft-robots/

A better way to control shape-shifting soft robots

Imagine a slime-like robot that can seamlessly change its shape to squeeze through narrow spaces, which could be deployed inside the human body to remove an unwanted item.

While such a robot does not yet exist outside a laboratory, researchers are working to develop reconfigurable soft robots for applications in health care, wearable devices, and industrial systems.

But how can one control a squishy robot that doesn’t have joints, limbs, or fingers that can be manipulated, and instead can drastically alter its entire shape at will? MIT researchers are working to answer that question.

They developed a control algorithm that can autonomously learn how to move, stretch, and shape a reconfigurable robot to complete a specific task, even when that task requires the robot to change its morphology multiple times. The team also built a simulator to test control algorithms for deformable soft robots on a series of challenging, shape-changing tasks.

Play video

Their method completed each of the eight tasks they evaluated while outperforming other algorithms. The technique worked especially well on multifaceted tasks. For instance, in one test, the robot had to reduce its height while growing two tiny legs to squeeze through a narrow pipe, and then un-grow those legs and extend its torso to open the pipe’s lid.

While reconfigurable soft robots are still in their infancy, such a technique could someday enable general-purpose robots that can adapt their shapes to accomplish diverse tasks.

“When people think about soft robots, they tend to think about robots that are elastic, but return to their original shape. Our robot is like slime and can actually change its morphology. It is very striking that our method worked so well because we are dealing with something very new,” says Boyuan Chen, an electrical engineering and computer science (EECS) graduate student and co-author of a paper on this approach.

Chen’s co-authors include lead author Suning Huang, an undergraduate student at Tsinghua University in China who completed this work while a visiting student at MIT; Huazhe Xu, an assistant professor at Tsinghua University; and senior author Vincent Sitzmann, an assistant professor of EECS at MIT who leads the Scene Representation Group in the Computer Science and Artificial Intelligence Laboratory. The research will be presented at the International Conference on Learning Representations.

Controlling dynamic motion

Scientists often teach robots to complete tasks using a machine-learning approach known as reinforcement learning, which is a trial-and-error process in which the robot is rewarded for actions that move it closer to a goal.

This can be effective when the robot’s moving parts are consistent and well-defined, like a gripper with three fingers. With a robotic gripper, a reinforcement learning algorithm might move one finger slightly, learning by trial and error whether that motion earns it a reward. Then it would move on to the next finger, and so on.

But shape-shifting robots, which are controlled by magnetic fields, can dynamically squish, bend, or elongate their entire bodies.

The researchers built a simulator to test control algorithms for deformable soft robots on a series of challenging, shape-changing tasks. Here, a reconfigurable robot learns to elongate and curve its soft body to weave around obstacles and reach a target.

Image: Courtesy of the researchers

“Such a robot could have thousands of small pieces of muscle to control, so it is very hard to learn in a traditional way,” says Chen.

To solve this problem, he and his collaborators had to think about it differently. Rather than moving each tiny muscle individually, their reinforcement learning algorithm begins by learning to control groups of adjacent muscles that work together.

Then, after the algorithm has explored the space of possible actions by focusing on groups of muscles, it drills down into finer detail to optimize the policy, or action plan, it has learned. In this way, the control algorithm follows a coarse-to-fine methodology.

“Coarse-to-fine means that when you take a random action, that random action is likely to make a difference. The change in the outcome is likely very significant because you coarsely control several muscles at the same time,” Sitzmann says.

To enable this, the researchers treat a robot’s action space, or how it can move in a certain area, like an image.

Their machine-learning model uses images of the robot’s environment to generate a 2D action space, which includes the robot and the area around it. They simulate robot motion using what is known as the material-point-method, where the action space is covered by points, like image pixels, and overlayed with a grid.

The same way nearby pixels in an image are related (like the pixels that form a tree in a photo), they built their algorithm to understand that nearby action points have stronger correlations. Points around the robot’s “shoulder” will move similarly when it changes shape, while points on the robot’s “leg” will also move similarly, but in a different way than those on the “shoulder.”

In addition, the researchers use the same machine-learning model to look at the environment and predict the actions the robot should take, which makes it more efficient.

Building a simulator

After developing this approach, the researchers needed a way to test it, so they created a simulation environment called DittoGym.

DittoGym features eight tasks that evaluate a reconfigurable robot’s ability to dynamically change shape. In one, the robot must elongate and curve its body so it can weave around obstacles to reach a target point. In another, it must change its shape to mimic letters of the alphabet.

In this simulation, the reconfigurable soft robot, trained using the researchers’ control algorithm, must change its shape to mimic objects, like stars, and the letters M-I-T.

Image: Courtesy of the researchers

“Our task selection in DittoGym follows both generic reinforcement learning benchmark design principles and the specific needs of reconfigurable robots. Each task is designed to represent certain properties that we deem important, such as the capability to navigate through long-horizon explorations, the ability to analyze the environment, and interact with external objects,” Huang says. “We believe they together can give users a comprehensive understanding of the flexibility of reconfigurable robots and the effectiveness of our reinforcement learning scheme.”

Their algorithm outperformed baseline methods and was the only technique suitable for completing multistage tasks that required several shape changes.

“We have a stronger correlation between action points that are closer to each other, and I think that is key to making this work so well,” says Chen.

While it may be many years before shape-shifting robots are deployed in the real world, Chen and his collaborators hope their work inspires other scientists not only to study reconfigurable soft robots but also to think about leveraging 2D action spaces for other complex control problems.