Univ. of California, Berkeley researchers have developed algorithms that enable robots to learn motor tasks through trial and error using a process that more closely approximates the way humans learn, marking a major milestone in the field of artificial intelligence. They demonstrated their technique, a type of reinforcement learning, by having a robot complete various tasks without pre-programmed details about its surroundings.
Using a smart tablet and a red beam of light, Georgia Institute of Technology researchers have...
Robotic dogs are likely to replace the real thing in households worldwide in as little as a...
To make cars as safe as possible, we crash them into walls to pinpoint weaknesses and better...
For the last decade, scientists have deployed increasingly capable underwater robots to map and monitor pockets of the ocean to track the health of fisheries, and survey marine habitats and species. In general, such robots are effective at carrying out low-level tasks, specifically assigned to them by human engineers, a tedious and time-consuming process for the engineers.
For decades, robots have advanced the efficiency of human activity. Typically, however, robots are formed from bulky, stiff materials and require connections to external power sources; these features limit their dexterity and mobility. But what if a new material would allow for development of a "soft robot" that could reconfigure its own shape and move using its own internally generated power?
Most people are naturally adept at reading facial expressions to tell what others are feeling. Now scientists have developed ultra-sensitive, wearable sensors that can do the same thing. Their technology, reported in the ACS Nano, could help robot developers make their machines more human.
In a rolling, outdoor field, full of lumps, bumps and uneven terrain, researchers at Oregon State University last week successfully field-tested for the first time the locomotion abilities of a two-legged robot with technology that they believe heralds the running robots of the future.
A Univ. of Sydney researcher has designed and successfully tested a method for autonomously docking drones for refueling or recharging, in mid-air. He used a combination of precise measurements from an infrared camera, with GPS and inertial sensors to allow the sky-high docking to occur.
A first-of-its-kind robotic vehicle recently dove to depths never before visited under Antarctica’s Ross Ice Shelf and brought back video of life on the seafloor. A team of scientists and engineers from the Georgia Institute of Technology assembled the unmanned, underwater vehicle on Antarctica. They deployed (and retrieved) the vehicle through a 12-in diameter hole through 20 m of ice and another 500 m of water to the sea floor.
Dielectric elastomers are novel materials for making actuators or motors with soft and lightweight properties that can undergo large active deformations with high-energy conversion efficiencies. This has made dielectric elastomers popular for creating devices such as robotic hands, soft robots, tunable lenses and pneumatic valves, and possibly flapping robotic wings.
Researchers at Carnegie Mellon Univ. (CMU) who develop snake-like robots have picked up a few tricks from real sidewinder rattlesnakes on how to make rapid and even sharp turns with their undulating, modular device. Working with colleagues at the Georgia Institute of Technology and Zoo Atlanta, they have analyzed the motions of sidewinders and tested their observations on CMU’s snake robots.
The most realistic risks about the dangers of artificial intelligence are basic mistakes, breakdowns and cyber attacks, an expert in the field says—more so than machines that become super powerful, run amok and try to destroy the human race.
Researchers building a new underwater robot they’ve dubbed the “Millennium Falcon” certainly have reason to believe it will live up to its name. The robot will deploy instruments to gather information in unprecedented detail about how marine life interacts with underwater equipment used to harvest wave and tidal energy.
People typically consider doing the laundry to be a boring chore. But laundry is far from boring for artificial intelligence (AI) researchers. To AI experts, programming a robot to do the laundry represents a challenging planning problem because current sensing and manipulation technology is not good enough to identify precisely the number of clothing pieces that are in a pile and the number that are picked up with each grasp.
Scientists have developed an octopus-like robot, which can zoom through water with ultra-fast propulsion and acceleration never before seen in man-made underwater vehicles. Most fast aquatic animals are sleek and slender to help them move easily through the water but cephalopods, such as the octopus, are capable of high-speed escapes by filling their bodies with water and then quickly expelling it to dart away.
When disaster strikes, it's important for responders and emergency officials to know what critical infrastructure has been damaged so they can direct supplies and resources accordingly. Doug Stow, a geography professor from San Diego State Univ., is developing a program that uses before-and-after aerial imagery to reveal infrastructure damage in a matter of minutes.
Many people imagine robots today as clunky, metal versions of humans, but scientists are forging new territory in the field of “soft robotics”. One of the latest advances is a flexible, microscopic hand-like gripper. The development could help doctors perform remotely guided surgical procedures or perform biopsies. The materials also could someday deliver therapeutic drugs to hard-to-reach places.
The editors of R&D Magazine are looking for speakers to participate in a webinar on “Using Multiple Materials in 3D Printing.” Candidates are asked to give a 15-min PowerPoint-based talk over the phone on their experiences in fabricating 3D printed products using multiple materials or developing the processes and/or technologies to accomplish this.
Grasping an object involves a complex network of brain functions. First, visual cues are processed in specialized areas of the brain. Then, other areas of the brain use these signals to control the hands to reach for and manipulate the desired object. New findings suggest that the cerebellum may play a critical role. Findings could lead to advancements in assistive technologies benefiting the disabled.
A device, possibly an unmanned aerial drone, was found on the White House grounds during the middle of the night while President Barack Obama and the first lady were in India, but his spokesman said today that it posed no threat. It was unclear whether their daughters, Sasha and Malia, were at home at the time of the incident with their grandmother, Marian Robinson, who also lives at the White House.
A new cardboard-robotic toolkit allows children to create custom robots they control wirelessly with hand gestures without formal education in programming or electronics. The system, called HandiMate, uses motorized "joint modules" equipped with wireless communicators and microcontrollers. Children create robots by using Velcro strips to attach the modules to any number of everyday materials and objects.
The stars are aligning for science and engineering, as a new movie about a high school robotics team makes its debut in theaters nationwide. The movie, “Spare Parts,” is based on FIRST Robotics Competition Team 842 - Falcon Robotics, from Carl Hayden Community High School in Phoenix, Ariz., and their famous robotic underdog victory against MIT which was chronicled in the WIRED article “La Vida Robot” in 2005.
For household robots ever to be practical, they’ll need to be able to recognize the objects they’re supposed to manipulate. But while object recognition is a highly studied topic in artificial intelligence, even the best object detectors still fail much of the time. Researchers at MIT believe that household robots should take advantage of their mobility and their relatively static environments to make object recognition easier.
The Federal Aviation Administration has issued permits to use drones to monitor crops and photograph properties for sale, marking the first time permission has been granted to companies involved in agriculture and real estate. The exemptions to the current ban on commercial drone flights were granted to Advanced Aviation Solutions in Star, Idaho, for “crop scouting,” and to Douglas Trudeau of Tierra Antigua Realty in Tucson, Arizona.
Sandia National Laboratories is tackling one of the biggest barriers to the use of robots in emergency response: energy efficiency. Through a project supported by DARPA, Sandia is developing technology that will dramatically improve the endurance of legged robots, helping them operate for long periods while performing the types of locomotion most relevant to disaster response scenarios.
A Univ. of Texas at Dallas professor applied robot control theory to enable powered prosthetics to dynamically respond to the wearer’s environment and help amputees walk. In recently published research, wearers of the robotic leg could walk on a moving treadmill almost as fast as an able-bodied person.
Bathymetric lidars are used today primarily to map coastal waters. At nearly 600 lbs, the systems are large and heavy, and they require costly, piloted aircraft to carry them. A team at the Georgia Tech Research Institute has designed a new approach that could lead to bathymetric lidars that are much smaller and more efficient than the current full-size systems.
A National Science Foundation-funded research team has successfully tested an autonomous underwater vehicle (AUV) that can produce high-resolution, 3-D maps of Antarctic sea ice. SeaBED, as the vehicle is known, measured and mapped the underside of sea-ice floes in three areas off the Antarctic Peninsula that were previously inaccessible.
- Page 1