Articles tagged with Robots
Researchers have developed a robotic gripping mechanism inspired by the sea anemone's ability to catch prey. The device can grasp various objects of different sizes, shapes, and materials using its thermoplastic rubber skin.
A team of scientists has created the first living robots, 'xenobots', by assembling frog embryo cells into new life forms. These biodegradable organisms can move, heal themselves, and even carry payloads, promising advances in drug delivery, toxic waste clean-up, and more.
Researchers have developed bioinspired shapes that serve as acoustically conspicuous guideposts for sonar-guided autonomous vehicles. These bioinspired sonar reflectors were tested in experiments and showed promising results, enabling robots to navigate through new environments with improved accuracy.
The MIPT team developed a new method for reinforcement learning with demonstrations, enabling rapid solution of hierarchical problems in Minecraft. This approach opens opportunities for applying reinforcement learning to real-world tasks like autonomous vehicle control and manipulator control.
A team of Brown University researchers has developed a way to measure the forces involved in the Cheerios effect, a phenomenon where cereal bits cluster together in milk. The experiment revealed that traditional models under-predicted the force, but adding tilt conditions improved agreement.
The 5G-DIVE project validates 5G technology for remote control of industrial robots and autonomous coordinated drone flight, requiring high bandwidth and low latency. The project aims to improve performance and corporate value through Fog Computing and intelligent design.
Researchers are working on a new project to develop transparent and explainable deep-learning AI computing models that make decisions. The project will use adversarial training techniques to understand how neural networks learn and make predictions.
Nader Motee is investigating real-time perception and planning for networks of robots, aiming to improve efficiency and resiliency. His research also focuses on risk-aware planning and control to mitigate effects of local failures in nonlinear dynamical networks.
Researchers are developing robotic networks that can work independently but collaboratively to accomplish complex tasks. By using genetic fuzzy logic, they were able to get three robots and then five robots to move a token to a designated spot on a table.
Skoltech PhD students won the Best Demo Award at ACM Siggraph Asia 2019 for their SwarmCloak technology, a novel system for landing flying robots onto the human arm. The team successfully demonstrated the tech over three days with no malfunctions.
The NUS researchers developed a novel metal-based material using platinum and burnt paper, which is half as light as paper and can withstand temperatures up to 800°C. This material enables the creation of flexible and lightweight prosthetic limbs with real-time strain sensing capabilities.
A US Army Research Laboratory team developed new algorithms to integrate context into artificial intelligence, advancing robotics and AI processes in areas like natural language communication and world model development. The research supports the Next Generation Combat Vehicle by integrating context-driven AI within human-autonomy teams.
Researchers at MIT developed a model that learns a compact state representation for soft robots, optimizing movement control and material design parameters. This enables 2D and 3D soft robots to complete tasks quickly and accurately in simulations.
A new robotic skin called ElectroSkin has been created, which can crawl across surfaces using artificial muscles and electrical charges. This innovative technology could lead to the development of soft robots for environmental monitoring, robot grippers, and wearable technologies.
Tiny robots equipped with magnets are steered through arteries to break up blood clots using controlled rotating magnetic fields. This non-invasive method has the potential to improve patient comfort and reduce the risk of infection, making it a promising alternative to current treatments.
MM3D printing breaks the speed barrier for multimaterial 3D printing by switching between up to eight materials at 50 times per second. This enables the creation of complex shapes and origami-like architectures with high quality transitions.
The new SLAS Technology auto-commentary discusses the challenges of gathering analytical measurements and introduces realized-based measurement systems as potential replacements. These systems use robots as central system integrators, providing flexibility and enabling decentralized open systems.
Researchers argue that incorporating sex and gender into scientific studies can improve experiments, reduce bias, and lead to new discoveries. Examples include advancements in climate change research, facial recognition systems, and social robots. The goal is to increase transparency and promote inclusion in research.
A new USC study found that training robots with a human adversary significantly improves their grasp of objects, leading to better performance in manipulation tasks. The researchers discovered that this approach helps robots learn more stable grasps and succeed more often in real-world environments.
Researchers create method that enables robots to use environmental clues to plan routes, reducing time spent exploring properties and eliminating the need for maps of specific residences. The approach leverages pre-existing algorithms to generate a new map of the environment as the robot moves around, represented as semantic clues.
Cornell researchers develop inexpensive, touch-sensitive soft robots to predict grape yield and detect fungal threats. The method uses machine learning to analyze plant growth patterns, allowing for early prediction and more time for planning.
Researchers at Purdue University have developed fast-acting robots capable of re-creating bioinspired high-powered and high-speed motions using stored elastic energy. These robots, similar to the chameleon's tongue strike, can catch live insects in just 120 milliseconds.
A new MIT system uses computer-vision techniques to detect and classify subtle changes in shadows on the ground, enabling autonomous vehicles to quickly avoid collisions with other objects. The system outperforms traditional LiDAR technology by more than half a second in real-world scenarios.
A joint research team has successfully designed a swarm of tiny drones to explore unknown environments autonomously. The drones use a novel bug algorithm that enables them to navigate back to a base station while avoiding obstacles and detecting victims, showcasing the potential for swarms in search-and-rescue scenarios.
Researchers at MIT have compiled a comprehensive dataset of robotic pushing dynamics, capturing the behavior of hundreds of different objects. This dataset, called Omnipush, enables robots to learn fundamental object manipulation tasks, such as reorienting and inspecting objects, by training on a diverse range of pushing scenarios.
Researchers developed a new technology to produce flexible 3D medical devices by selectively bonding polymeric thin films using plasma. This method overcomes limitations of existing flexible 3D structures, enabling mass production with customized shapes and wire patterns.
Researchers at MIT have created an algorithm that significantly speeds up the planning process required for robots to adjust their grasp on objects. The new approach uses motion cones to efficiently calculate feasible pushes and reposition objects in less than a second, compared to traditional algorithms that take over 500 seconds.
Researchers have developed a new kind of robotics that can assemble large structures from identical subunits using simple robotic systems. The system works by having the robot work together with the structure, adjusting its position as it adds each piece, allowing for efficient and precise assembly.
A new study examines the effectiveness of artificial intelligence in mediating conflict among team members. Researchers found that while humans initially engaged with a virtual agent during mission planning, their engagement decreased as conflicts arose, suggesting a limit to trust in AI-mediated conflict resolution.
Researchers at UC San Diego developed soft actuators that can be controlled electrically, making them compatible with small electronic devices and batteries. These actuators enabled the creation of compact, portable and multifunctional soft robots with various applications.
The research team aims to design algorithms that optimally distribute computation between robots and the cloud for guaranteed safe robotic operation. 5G systems offer higher data rates but pose challenges due to limited bandwidth and high latency, requiring new control algorithms that can exploit high-rate links when available.
Researchers at U.S. Army Research Laboratory have made significant breakthroughs in developing artificial nanomotors inspired by biological molecules, which can harness Brownian motion for efficient energy production. These advancements aim to create faster, more versatile robots with improved autonomy and stealth capabilities.
Deep3DFly uses computer science and neuroscience to model fly movements and poses, enabling robots to climb challenging terrain. The system can automatically predict and correct errors in calculations, opening up new possibilities for robotics and medicine.
The CURSOR project aims to locate survivors trapped under rubble more swiftly and with less risk. The EU-Japan collaboration has developed miniaturized robots equipped with cameras, thermographic capabilities, and chemical sensors to detect human presence.
Researchers created a squidlike robot that uses pulsed jet propulsion, achieving high speeds while minimizing environmental impact. The device can test water samples simultaneously, making it suitable for environmental monitoring.
Compound lenses developed by researchers are inspired by mosquito eyes, providing a wide field of view with rapid imaging capabilities. The new design has antifogging properties, similar to those found in mosquito eyes, making it suitable for applications in drones, robots and medical devices.
A new study by the US Army Research Laboratory found that humans' confidence in robots decreases significantly after a mistake, regardless of transparency. Participants rated the unreliable robot as less trustworthy, likable, and safe than the reliable one.
Researchers created robots entirely from smaller robots called smarticles, unlocking a new locomotion technique. The smarticles can form structures and move in response to stimuli, allowing them to navigate mazes and even change their shapes.
Researchers developed Tunabot to better understand fish propulsion, which could lead to faster, more efficient propulsion systems for underwater vehicles. The robot's design was informed by studies of yellowfin tuna and mackerel, and its performance data sets a high standard for the field of fish robotics.
Researchers develop robotic fish predators that induce fear responses in mosquitofish, triggering avoidance behaviors and physiological changes associated with energy reserve loss. Brief exposure to a robotic largemouth bass replica provokes meaningful stress responses in mosquitofish.
The SUTD researchers developed a strong, flexible and scalable multi-robots system that can be used in intelligence, surveillance and reconnaissance operations. The system includes a combination of a wheeled ground unit and miniature robots, allowing it to be scalable and explore dynamic environments.
Researchers at Cornell University have developed a stretchable optical lace material that enables soft robots to sense their environment through mechanosensors. The material allows robots to detect changes in stress and strain, enabling them to adjust their actions accordingly.
Researchers at Carnegie Mellon University have created a computer model that can translate text describing physical movements into simple computer-generated animations. The Joint Language-to-Pose model enables sentences and physical motions to be jointly embedded, learning how language is related to action, gestures, and movement.
Researchers found that immediate robot responses stimulated wild Galápagos lizards to react more quickly and often than delayed responses. This may help lizards assess their competitors' aggression levels and avoid injury. The study used realistic robots to simulate displays, providing new insights into lizard communication.
Researchers from SUTD, SUSTech, and ZJU proposed a generic process flow for guiding DLP 3D printing of miniature pneumatic actuators for soft robots. This enables the production of miniature soft robotic actuators with complex geometries and sophisticated multimaterial designs.
A team of scientists from FEFU developed a software module to automatically diagnose defects in sensors and electric drives in various kinds of robots. The system can compensate for detected defects in real time, preserving the required levels of operating quality.
Biomechanics researchers at Georgia Institute of Technology used cockroaches' sprints to develop a method for assessing and improving robot locomotion. The new approach focuses on phase-coupling oscillations, allowing it to work with both insect and robotic systems.
Researchers developed self-folding soft robots inspired by origami, using 3D-printed active hinges that can be programmed to fold at different temperatures. The Rollbot, a flat sheet that curls into a wheel and propels itself, demonstrates the method's capabilities.
Researchers at the University of Washington have developed an algorithm that can monitor factory and warehouse workers' behavior in real-time, providing risk scores and suggestions for improving ergonomics. The app will alert workers to high-risk actions and provide feedback on how to modify their behaviors.
Researchers have developed a soft and stretchable pump, eliminating tethers in soft robots. The pump uses an electrical field to circulate liquids, promising applications in exoskeletons, robotic clothing, and smart clothing.
Researchers argue that robots must understand the reason behind their actions to work alongside people safely and effectively. This shift in thinking is necessary as automation increases and human-robot interaction grows.
A study published in Nature Human Behaviour found that people are more upset about losing their jobs to robots than to other workers. The researchers discovered that this preference reverses when it comes to one's own job, with most individuals finding robot replacement less threatening to their self-worth.
The project aims to determine the forces that dictate underwater walking, shedding light on the colonization of land by animals. By studying Spanish ribbed newts, researchers hope to gain insight into the biomechanics of animal movement.
Scientists at North Carolina State University have created a method to remotely control, lock into place, and later transform the shape of soft robots using light and magnetic fields. The technology is promising for medical and aerospace industries.
Scientists at Huazhong University of Science & Technology have created a bio-inspired untethered fully soft robot in liquid that can actuate using environmental energy gradients. The robot achieves an impressive speed of 7 times higher than the best reported value for untethered soft robotic fish.
Boston University researchers found specialized brain cells in rats that provide personal maps of their surroundings. These cells could help create smarter autonomous vehicles by allowing them to navigate complex environments like humans do.
A computational simulation suggests that insects can use skylight properties to determine their compass direction with an error of less than two degrees. This discovery could inspire designs for new navigation tools for robots, potentially replacing GPS technology.
The Asynchronous Coded Electronic Skin (ACES) system can detect touches 1,000 times faster than the human sensory nervous system. It achieves ultra-high responsiveness and robustness to damage, making it suitable for robots and prosthetic devices.
Scientists developed a command-and-control plugin for intelligent industrial robots, allowing for high-quality 3D computer models to be built quickly and precisely. The software helps solve the issue of hard programming of industrial robots and can fix in-process workpieces on universal positioning devices.
Researchers at Georgia Institute of Technology have developed micro-bristle-bots that harness vibration to move and interact with their environment. The bots can be controlled by adjusting vibration frequencies and can potentially be used for tasks such as repairing injuries inside the human body or sensing environmental changes.