Articles tagged with Robots
Researchers created tiny building blocks called voxels that exhibit special properties and can be assembled into large, complex objects. Examples include cars, robots, and wind turbine blades that respond to environmental stimuli in predictable ways.
A new USC study reveals that AI models lack common sense to generate plausible sentences, despite advances in natural language processing. The research challenges the effectiveness of current benchmark tests and finds that even the strongest models can make silly mistakes.
Researchers at the University of California, Berkeley, have created AI software that gives robots speed and skill to grasp objects, making it feasible for them to assist humans in warehouses. The technology reduces computation time from 29 seconds to under one-tenth of a second.
Researchers developed machine learning frameworks that guarantee robots' performance in unfamiliar settings, with a guaranteed success rate of 88.4% in obstacle avoidance trials. The approach expands generalization theory to robotics, providing more broadly applicable guarantees on robot control policies.
A new paper argues that robots designed for older adults can provide essential human values like social interaction, emotional fulfillment, and bodily integrity. The author suggests that Western cultures' narrow view of sex as lust needs to be challenged, and that robots can enable people to experience physical affection and self-respect.
Designing and marketing sexbots for older adults with disabilities could promote emotional health and wellbeing, counter ageism and stereotyping. This is based on research showing that sexual desires are fundamental to human dignity and respect, even in older age.
Researchers at Michigan State University are developing softer materials for robots to navigate safely and build trust with humans. The team plans to test their 'Soft Mult-Arm Robot' prototype in apple orchards and operating rooms.
Researchers from the University of Tsukuba found that praising participants with robots and virtual agents improved offline motor task performance. Two agents outperformed one agent in terms of skill consolidation.
Researchers discovered that squid jet propulsion can be more efficient when considering turbulent flow conditions, revealing clues about how squids maneuver within these environments. The study found symmetry-breaking instability of vortices around the jetter, which affects thrust production and efficiency.
Fish schools are highly dynamic, social systems. Researchers used biomimetic robots to show that fish can profit from the vortices generated by near neighbours without keeping fixed distances, and found that a specific time lag based on spatial position is key to energy efficiency.
A team of researchers at the University of Illinois created a new take on deep reinforcement learning using the game Capture the Flag, helping robots evaluate their next move and adapt to unexpected situations. By breaking down tasks into sub-tasks, they improved adaptation and reduced complexity in updates.
Researchers at Princeton University developed a unique installation, LightVault, using robotic strength and precision to reduce resource use. The structure's doubly curved design improved its structural efficiency by reducing material requirements.
Researchers at NTU Singapore have developed a system that enables robots to recognize pain and self-repair with minimal wiring and circuits. The 'mini-brain' approach allows the robot to process information efficiently and adapt to its environment, making it more interactive with humans.
Scientists use a custom robot to survey how mutations in regulatory regions of the genome affect animal development, revealing that most mutations alter gene expression in some way. The study finds that regulatory regions encode valuable information densely and that single mutations can have several different effects.
A Purdue University team has created a mobile docking system for AUVs, enabling them to perform longer tasks without human intervention. The system uses algorithms to optimize trajectories, allowing robots to autonomously dock mid-mission to recharge and transfer data.
MoonRanger, a small robotic rover developed by Carnegie Mellon University, will search for signs of water at the moon's south pole. The rover is designed to explore at unprecedented speeds in both sunlit and dark conditions, and will use innovative technologies such as laser line stripes for night vision.
Researchers at Ohio State University have developed soft robots that can be controlled by magnetic fields, allowing for faster and less invasive delivery of medications. The 'soft' component of the robot is crucial, as it eliminates the need for motors, controllers, and tethers.
Researchers found that attributing a mind to an android and then 'dehumanizing' it leads to the uncanny valley phenomenon. The study suggests that at first sight humans anthropomorphize an android, but within milliseconds they detect deviations and dehumanize it.
A new study found that robotic support pets for older adults and people with dementia acquire bacteria over time, posing a risk of illness. However, a simple cleaning procedure involving anti-bacterial products and wipes can effectively reduce microbial loads to safe levels.
The new microscopic robots are controlled by flashing laser pulses and can walk thanks to a simple circuit made from silicon photovoltaics. The researchers explored ways to enhance the robots with more complicated electronics and onboard computation, potentially leading to swarms of tiny robots crawling through human tissue and blood.
A new study reveals how remora suckerfish detach themselves from surfaces and explores the application of this mechanism in underwater robots. The research team developed a biomimetic flexible adhesive disc with controllable motion, which showed similar detachment motions to its biological counterpart.
Researchers have developed a new rechargeable zinc battery that integrates into the structure of a robot to provide much more energy. The battery uses a network of aramid nanofibers and a water-based polymer gel, making it environmentally friendly and efficient.
Experts say AI's portrayal as White removes people of color from the future, perpetuating racial biases in algorithms and workforce. Cultural depictions of AI as White risk creating a 'racially homogenous' technologist workforce.
University of Michigan engineers have created centimeter-sized robots capable of more than ever before, leveraging origami principles to enhance their speed, agility, and control. The microbots can form complex shapes, complete tasks, and reconfigure into new shapes up to 80 times per second.
Researchers at IIT and Harvard University identified the tricks used by the brain to discriminate and generalize odorous molecules. The study sheds light on how our brain processes information from the sense of smell, revealing a new mechanism for recognizing specific smells.
Researchers developed a single-camera machine vision algorithm allowing indoor robots to guide themselves by identifying reference points on a tiled floor. The technology has wide-ranging potential applications in warehouses, distribution centers, and industrial settings.
A Carnegie Mellon-developed navigation system, SemExp, uses machine learning to train robots to recognize objects and understand their locations in a home. This enables the system to think strategically about how to search for something, making it more efficient than classical robotic navigation systems.
The NUS team created a sensory integrated artificial brain system that mimics biological neural networks, combining artificial skin and vision sensors to enable accurate conclusions about objects in real-time. The system achieved over 92% accuracy in classifying Braille letters and outperformed traditional systems in grasping tasks.
Researchers create 3D Dynamic Scene Graphs, enabling robots to quickly generate maps of their surroundings and extract relevant information. The new model is modeled after human perception and allows robots to navigate and make decisions like humans.
Researchers at Carnegie Mellon University developed a system that enables robots to pick up transparent and reflective objects using a color camera and depth images. The system outperforms previous methods in grasping these challenging objects, with high success rates in both clean and cluttered environments.
The study used advanced metrics from information theory to describe the collective behavior of small schools of ayu fish, finding that groups of three or more exhibited noticeably different dynamics. This work may aid in developing cooperative biomimetic swarming robots and help understand fundamental problems in complexity theory.
Researchers propose STyLuS*, a new approach to solving complex robot control problems, by growing 'trees' in the search space to find optimal solutions. This method reduces memory requirements and processing power needed for large-scale robotic tasks.
Researchers created jellyfish-bots that can outswim real-life counterparts and exhibit powerful movement. The new technique uses pre-stressed polymers to make soft robots more efficient, enabling faster speeds and improved performance.
Using highly novel techniques, researchers observed upper ocean currents of approximately 0.1-10 km in size, critical for controlling heat and carbon movement between the ocean and atmosphere. These findings contribute to improving our understanding of small-scale ocean and climate processes with impacts globally.
Researchers at UC Riverside developed a new film made of gold nanoparticles that can respond to any type of movement, enabling robots to mimic chameleons. The material's complex patterns can be displayed through programming, opening up various applications such as underwater exploration and authentication features.
Scientists study cockroach movements to understand how animals overcome energy barriers in tricky terrain. By mimicking cockroaches' 'jittering' legs, researchers create robots that can transition between movements with ease.
Researchers created flexible feet using coffee grounds and structures inspired by nature, allowing robots to walk up to 40% faster and grip better on various terrains. The new design outperforms traditional robot feet in adapting to natural environments.
Researchers from National University of Singapore developed a new stretchable material called HELIOS that can store more electronic charges at lower voltages, enabling higher brightness and longer operating lifetime. The material has self-healing properties, allowing it to repair itself under ambient environmental conditions.
Researchers demonstrate that locally observed robot distribution can correlate with environmental features, such as exits in office-like environments. This approach enables trapped office workers to navigate their way out of a collapsed building, even in scenarios where robots lack communication or sensors.
Researchers at Harvard have developed a fast-moving jumping soft actuator that harnesses the energy released by buckling to achieve speed. The device uses shell buckling similar to toy poppers, enabling it to catapult itself into the air and navigate safely through uncharted landscapes.
Christoph Keplinger's research focuses on soft robotics, artificial muscles, and medical applications. He aims to rethink robotics by merging soft matter with advanced technologies.
Researchers found that audiences preferred robot comedians with good timing, leading to higher laughter ratings. Adaptive performances also improved audience perception of individual jokes, but not overall funniness.
Researchers developed soft robotic technology using reservoir computing, enabling flexible control of pneumatic artificial muscles. This innovation may lead to wearable rehabilitation devices, biomedical robots, and remote-sensing applications, all with improved safety and adaptability.
A group of researchers developed a new way for robots to pool data in real-time, allowing them to navigate difficult terrain as a team. The system uses a centralized data cloud, where each robot draws on data from other robots to steer clear of obstacles.
Researchers developed a real-time physics engine for simulating the movements of soft robots, which can be used for autonomous operations. The engine uses the discrete elastic rods (DER) algorithm, allowing for faster-than-real-time simulation of hundreds of movements.
The new soft robotics, called LEAP, have a spring-powered spine that enables rapid switching between stable states, allowing for faster movement and more forceful grasping. This innovation outperforms previous soft robots by reaching speeds of up to three times faster on solid surfaces.
Researchers studied geckos' unique toe mechanism, which allows them to adjust and distribute weight on various surfaces. This adaptation enables agile locomotion and is crucial for designing effective robot feet and grippers.
A new research study investigated how service robots in hotels could help redefine leadership and boost the hospitality industry. The results showed that while service robots increase efficiency, they also pose challenges like high costs and skill deficits.
Engineers have demonstrated a technology that can transmit electricity wirelessly to recharge electric cars, robots or even drones while they move. The system boosts efficiency to 92% and can transmit 10 watts of electricity over short distances.
Cognitive neuroscientists are using computers to illuminate human emotions, overturning old ideas about emotion structure. Facial expression recognition across cultures reveals only four cross-cultural expressions, while digital agents can now generate nuanced, culturally sensitive expressions.
A study by MIT economist Daron Acemoglu finds that adding one robot per 1,000 workers reduces national employment-to-population ratio by about 0.2 percent, with some areas affected more than others. This translates to each additional robot replacing about 3.3 workers nationally, on average.
Researchers at the University of Illinois at Urbana-Champaign developed spinobots that can walk using a rat muscle and spinal cord tissue. The integration of the spinal cord gives them a natural walking rhythm, which could have implications for neurocomputing and restorative medicine.
Researchers have developed a metal-air scavenger that provides power by breaking chemical bonds in metal surfaces, outperforming batteries and harvesters. The technology has 10 times more power density than the best energy harvesters and 13 times more energy density than lithium-ion batteries.
A study published in Royal Society Open Science found that small and slowly moving robots can effectively monitor the behavior of salmon in fish farms without scaring them off. The results suggest that these robots can provide valuable insights into the state of the fish, including their health, parasites, and feeding habits.
A study published in the Journal of Economic Psychology found that people extend similar levels of trust to humans and robots. However, social emotions such as anger, gratitude, guilt, and pride vary depending on partner type, with participants experiencing more intense emotions when interacting with humans.
Researchers are using a new variant of reinforcement learning to train robots to predict multiple possible outcomes and choose the safest option. The algorithm will assign points to different outcomes, such as rewarding safer options and penalizing those that put the robot in danger.
William Beksi, a UTA computer scientist, is developing topological methods to process large-scale 3D point cloud data, aiming to overcome limitations in robotic perception applications. This new approach uses persistent homology to extract unique insights into the structure of point cloud data.
A new review examines two decades of research on how people develop trust in AI, finding representation, capabilities, and anthropomorphism play key roles. The study proposes a framework to guide organizations integrating AI into work.
Researchers found that robotic turtles with small size and slow movements disturb fish minimally, promoting better fish welfare and efficiency. This technology can provide online updates and monitoring without human intervention, leading to quicker responses and lower mortality rates.
Researchers outline robots' potential roles in clinical care, logistics, and reconnaissance to combat COVID-19. They suggest robots for disinfection, delivery, and monitoring compliance with quarantines.