Tag Archives: taught
#439443 This Robot Taught Itself to Run, Then ...
In the last few months, robots have learned some pretty cool new skills, including performing a sweet coordinated dance routine and making pizzas from start to finish. Now there’s another accomplishment to add to the list: a bipedal robot named Cassie just ran a 5K.
Made by Agility Robotics, which was spun out of Oregon State University, Cassie was developed using a $1 million grant from DARPA. The robot is basically a pair of mechanical legs with a battery pack sitting on top. Thanks to the design of its hip joints, its legs can move forward, backward, or side to side.
Earlier this year, a group of students at Berkeley used machine learning to teach Cassie to walk. But making the leap from walking to running wasn’t as straightforward as you might think. To us, running is just a faster version of walking, and we don’t often consider the various skills and brain regions that go into even a short jog around the neighborhood.
Our core muscles engage to help keep us balanced as we’re in constant motion. Our vision scans the area in front of us for obstacles to avoid, changing course as necessary. Our heart rate kicks up a few notches, and our respiratory system regulates our breathing.
Granted, it’s a little different for a robot, since they don’t have lungs or a heart. But they do have a “brain” (software), “muscles” (hardware), and “fuel” (a battery), and these all had to work together for Cassie to be able to run.
The brunt of the work fell to the brain—in this case, a machine learning algorithm developed by students at Oregon State University’s Dynamic Robotics Laboratory. Specifically, they used deep reinforcement learning, a method that mimics the way humans learn from experience by using a trial-and-error process guided by feedback and rewards. Over many repetitions, the algorithm uses this process to learn how to accomplish a set task. In this case, since it was trying to learn to run, it may have tried moving the robot’s legs varying distances or at distinct angles while keeping it upright.
Once Cassie got a good gait down, completing the 5K was as much a matter of battery life as running prowess. The robot covered the whole distance (a course circling around the university campus) on a single battery charge in just over 53 minutes, but that did include six and a half minutes of troubleshooting; the computer had to be reset after it overheated, as well as after Cassie fell during a high-speed turn. But hey, an overheated computer getting reset isn’t so different from a human runner pausing to douse their head and face with a cup of water to cool off, or chug some water to rehydrate.
Cassie isn’t the first two-legged robot to run. Honda’s Asimo robot had a slow jog down in 2004, and Boston Dynamics’ Atlas bot looks (sort of frighteningly) like a person when it runs, moving its arms in coordination with its legs. But it is notable that Cassie taught itself to run, as it shows off machine learning’s potential in robotic systems.
And this feat is just the beginning. “The students combined expertise from biomechanics and existing robot control approaches with new machine learning tools,” said Jonathan Hurst, a robotics professor who co-founded Agility in 2017. “This type of holistic approach will enable animal-like levels of performance. It’s incredibly exciting.”
Image Credit: Agility Robotics/Oregon State University Dynamic Robotics Laboratory Continue reading
#439105 This Robot Taught Itself to Walk in a ...
Recently, in a Berkeley lab, a robot called Cassie taught itself to walk, a little like a toddler might. Through trial and error, it learned to move in a simulated world. Then its handlers sent it strolling through a minefield of real-world tests to see how it’d fare.
And, as it turns out, it fared pretty damn well. With no further fine-tuning, the robot—which is basically just a pair of legs—was able to walk in all directions, squat down while walking, right itself when pushed off balance, and adjust to different kinds of surfaces.
It’s the first time a machine learning approach known as reinforcement learning has been so successfully applied in two-legged robots.
This likely isn’t the first robot video you’ve seen, nor the most polished.
For years, the internet has been enthralled by videos of robots doing far more than walking and regaining their balance. All that is table stakes these days. Boston Dynamics, the heavyweight champ of robot videos, regularly releases mind-blowing footage of robots doing parkour, back flips, and complex dance routines. At times, it can seem the world of iRobot is just around the corner.
This sense of awe is well-earned. Boston Dynamics is one of the world’s top makers of advanced robots.
But they still have to meticulously hand program and choreograph the movements of the robots in their videos. This is a powerful approach, and the Boston Dynamics team has done incredible things with it.
In real-world situations, however, robots need to be robust and resilient. They need to regularly deal with the unexpected, and no amount of choreography will do. Which is how, it’s hoped, machine learning can help.
Reinforcement learning has been most famously exploited by Alphabet’s DeepMind to train algorithms that thrash humans at some the most difficult games. Simplistically, it’s modeled on the way we learn. Touch the stove, get burned, don’t touch the damn thing again; say please, get a jelly bean, politely ask for another.
In Cassie’s case, the Berkeley team used reinforcement learning to train an algorithm to walk in a simulation. It’s not the first AI to learn to walk in this manner. But going from simulation to the real world doesn’t always translate.
Subtle differences between the two can (literally) trip up a fledgling robot as it tries out its sim skills for the first time.
To overcome this challenge, the researchers used two simulations instead of one. The first simulation, an open source training environment called MuJoCo, was where the algorithm drew upon a large library of possible movements and, through trial and error, learned to apply them. The second simulation, called Matlab SimMechanics, served as a low-stakes testing ground that more precisely matched real-world conditions.
Once the algorithm was good enough, it graduated to Cassie.
And amazingly, it didn’t need further polishing. Said another way, when it was born into the physical world—it knew how to walk just fine. In addition, it was also quite robust. The researchers write that two motors in Cassie’s knee malfunctioned during the experiment, but the robot was able to adjust and keep on trucking.
Other labs have been hard at work applying machine learning to robotics.
Last year Google used reinforcement learning to train a (simpler) four-legged robot. And OpenAI has used it with robotic arms. Boston Dynamics, too, will likely explore ways to augment their robots with machine learning. New approaches—like this one aimed at training multi-skilled robots or this one offering continuous learning beyond training—may also move the dial. It’s early yet, however, and there’s no telling when machine learning will exceed more traditional methods.
And in the meantime, Boston Dynamics bots are testing the commercial waters.
Still, robotics researchers, who were not part of the Berkeley team, think the approach is promising. Edward Johns, head of Imperial College London’s Robot Learning Lab, told MIT Technology Review, “This is one of the most successful examples I have seen.”
The Berkeley team hopes to build on that success by trying out “more dynamic and agile behaviors.” So, might a self-taught parkour-Cassie be headed our way? We’ll see.
Image Credit: University of California Berkeley Hybrid Robotics via YouTube Continue reading
#437687 Video Friday: Bittle Is a Palm-Sized ...
Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We’ll also be posting a weekly calendar of upcoming robotics events for the next few months; here's what we have so far (send us your events!):
ICRES 2020 – September 28-29, 2020 – Taipei, Taiwan
AUVSI EXPONENTIAL 2020 – October 5-8, 2020 – [Online]
IROS 2020 – October 25-29, 2020 – [Online]
CYBATHLON 2020 – November 13-14, 2020 – [Online]
ICSR 2020 – November 14-16, 2020 – Golden, Colo., USA
Let us know if you have suggestions for next week, and enjoy today's videos.
Rongzhong Li, who is responsible for the adorable robotic cat Nybble, has an updated and even more adorable quadruped that's more robust and agile but only costs around US $200 in kit form on Kickstarter.
Looks like the early bird options are sold out, but a full kit is a $225 pledge, for delivery in December.
[ Kickstarter ]
Thanks Rz!
I still maintain that Stickybot was one of the most elegantly designed robots ever.
[ Stanford ]
With the unpredictable health crisis of COVID-19 continuing to place high demands on hospitals, PAL Robotics have successfully completed testing of their delivery robots in Barcelona hospitals this summer. The TIAGo Delivery and TIAGo Conveyor robots were deployed in Hospital Municipal of Badalona and Hospital Clínic Barcelona following a winning proposal submitted to the European DIH-Hero project. Accerion sensors were integrated onto the TIAGo Delivery Robot and TIAGo Conveyor Robot for use in this project.
[ PAL Robotics ]
Energy Robotics, a leading developer of software solutions for mobile robots used in industrial applications, announced that its remote sensing and inspection solution for Boston Dynamics’s agile mobile robot Spot was successfully deployed at Merck’s thermal exhaust treatment plant at its headquarters in Darmstadt, Germany. Energy Robotics equipped Spot with sensor technology and remote supervision functions to support the inspection mission.
Combining Boston Dynamics’ intuitive controls, robotic intelligence and open interface with Energy Robotics’ control and autonomy software, user interface and encrypted cloud connection, Spot can be taught to autonomously perform a specific inspection round while being supervised remotely from anywhere with internet connectivity. Multiple cameras and industrial sensors enable the robot to find its way around while recording and transmitting information about the facility’s onsite equipment operations.
Spot reads the displays of gauges in its immediate vicinity and can also zoom in on distant objects using an externally-mounted optical zoom lens. In the thermal exhaust treatment facility, for instance, it monitors cooling water levels and notes whether condensation water has accumulated. Outside the facility, Spot monitors pipe bridges for anomalies.
Among the robot’s many abilities, it can detect defects of wires or the temperature of pump components using thermal imaging. The robot was put through its paces on a comprehensive course that tested its ability to handle special challenges such as climbing stairs, scaling embankments and walking over grating.
[ Energy Robotics ]
Thanks Stefan!
Boston Dynamics really should give Dr. Guero an Atlas just to see what he can do with it.
[ DrGuero ]
World's First Socially Distanced Birthday Party: Located in London, the robotic arm was piloted in real time to light the candles on the cake by the founder of Extend Robotics, Chang Liu, who was sat 50 miles away in Reading. Other team members in Manchester and Reading were also able to join in the celebration as the robot was used to accurately light the candles on the birthday cake.
[ Extend Robotics ]
The Robocon in-person competition was canceled this year, but check out Tokyo University's robots in action:
[ Robocon ]
Sphero has managed to pack an entire Sphero into a much smaller sphere.
[ Sphero ]
Squishy Robotics, a small business funded by the National Science Foundation (NSF), is developing mobile sensor robots for use in disaster rescue, remote monitoring, and space exploration. The shape-shifting, mobile, senor robots from UC-Berkeley spin-off Squishy Robotics can be dropped from airplanes or drones and can provide first responders with ground-based situational awareness during fires, hazardous materials (HazMat) release, and natural and man-made disasters.
[ Squishy Robotics ]
Meet Jasper, the small girl with big dreams to FLY. Created by UTS Animal Logic Academy in partnership with the Royal Australian Air Force to encourage girls to soar above the clouds. Jasper was created using a hybrid of traditional animation techniques and technology such as robotics and 3D printing. A KUKA QUANTEC robot is used during the film making to help the Australian Royal Airforce tell their story in a unique way. UTS adapted their High Accurate robot to film consistent paths, creating a video with physical sets and digital characters.
[ AU AF ]
Impressive what the Ghost Robotics V60 can do without any vision sensors on it.
[ Ghost Robotics ]
Is your job moving tiny amounts of liquid around? Would you rather be doing something else? ABB’s YuMi got you.
[ Yumi ]
For his PhD work at the Media Lab, Biomechatronics researcher Roman Stolyarov developed a terrain-adaptive control system for robotic leg prostheses. as a way to help people with amputations feel as able-bodied and mobile as possible, by allowing them to walk seamlessly regardless of the ground terrain.
[ MIT ]
This robot collects data on each cow when she enters to be milked. Milk samples and 3D photos can be taken to monitor the cow’s health status. The Ontario Dairy Research Centre in Elora, Ontario, is leading dairy innovation through education and collaboration. It is a state-of-the-art 175,000 square foot facility for discovery, learning and outreach. This centre is a partnership between the Agricultural Research Institute of Ontario, OMAFRA, the University of Guelph and the Ontario dairy industry.
[ University of Guleph ]
Australia has one of these now, should the rest of us panic?
[ Boeing ]
Daimler and Torc are developing Level 4 automated trucks for the real world. Here is a glimpse into our closed-course testing, routes on public highways in Virginia, and self-driving capabilities development. Our year of collaborating on the future of transportation culminated in the announcement of our new truck testing center in New Mexico.
[ Torc Robotics ] Continue reading
#437120 The New Indiana Jones? AI. Here’s How ...
Archaeologists have uncovered scores of long-abandoned settlements along coastal Madagascar that reveal environmental connections to modern-day communities. They have detected the nearly indiscernible bumps of earthen mounds left behind by prehistoric North American cultures. Still other researchers have mapped Bronze Age river systems in the Indus Valley, one of the cradles of civilization.
All of these recent discoveries are examples of landscape archaeology. They’re also examples of how artificial intelligence is helping scientists hunt for new archaeological digs on a scale and at a pace unimaginable even a decade ago.
“AI in archaeology has been increasing substantially over the past few years,” said Dylan Davis, a PhD candidate in the Department of Anthropology at Penn State University. “One of the major uses of AI in archaeology is for the detection of new archaeological sites.”
The near-ubiquitous availability of satellite data and other types of aerial imagery for many parts of the world has been both a boon and a bane to archaeologists. They can cover far more ground, but the job of manually mowing their way across digitized landscapes is still time-consuming and laborious. Machine learning algorithms offer a way to parse through complex data far more quickly.
AI Gives Archaeologists a Bird’s Eye View
Davis developed an automated algorithm for identifying large earthen and shell mounds built by native populations long before Europeans arrived with far-off visions of skyscrapers and superhighways in their eyes. The sites still hidden in places like the South Carolina wilderness contain a wealth of information about how people lived, even what they ate, and the ways they interacted with the local environment and other cultures.
In this particular case, the imagery comes from LiDAR, which uses light pulses that can penetrate tree canopies to map forest floors. The team taught the computer the shape, size, and texture characteristics of the mounds so it could identify potential sites from the digital 3D datasets that it analyzed.
“The process resulted in several thousand possible features that my colleagues and I checked by hand,” Davis told Singularity Hub. “While not entirely automated, this saved the equivalent of years of manual labor that would have been required for analyzing the whole LiDAR image by hand.”
In Madagascar—where Davis is studying human settlement history across the world’s fourth largest island over a timescale of millennia—he developed a predictive algorithm to help locate archaeological sites using freely available satellite imagery. His team was able to survey and identify more than 70 new archaeological sites—and potentially hundreds more—across an area of more than 1,000 square kilometers during the course of about a year.
Machines Learning From the Past Prepare Us for the Future
One impetus behind the rapid identification of archaeological sites is that many are under threat from climate change, such as coastal erosion from sea level rise, or other human impacts. Meanwhile, traditional archaeological approaches are expensive and laborious—serious handicaps in a race against time.
“It is imperative to record as many archaeological sites as we can in a short period of time. That is why AI and machine learning are useful for my research,” Davis said.
Studying the rise and fall of past civilizations can also teach modern humans a thing or two about how to grapple with these current challenges.
Researchers at the Institut Català d’Arqueologia Clàssica (ICAC) turned to machine-learning algorithms to reconstruct more than 20,000 kilometers of paleo-rivers along the Indus Valley civilization of what is now part of modern Pakistan and India. Such AI-powered mapping techniques wouldn’t be possible using satellite images alone.
That effort helped locate many previously unknown archaeological sites and unlocked new insights into those Bronze Age cultures. However, the analytics can also assist governments with important water resource management today, according to Hèctor A. Orengo Romeu, co-director of the Landscape Archaeology Research Group at ICAC.
“Our analyses can contribute to the forecasts of the evolution of aquifers in the area and provide valuable information on aspects such as the variability of agricultural productivity or the influence of climate change on the expansion of the Thar desert, in addition to providing cultural management tools to the government,” he said.
Leveraging AI for Language and Lots More
While landscape archaeology is one major application of AI in archaeology, it’s far from the only one. In 2000, only about a half-dozen scientific papers referred to the use of AI, according to the Web of Science, reputedly the world’s largest global citation database. Last year, more than 65 papers were published concerning the use of machine intelligence technologies in archaeology, with a significant uptick beginning in 2015.
AI methods, for instance, are being used to understand the chemical makeup of artifacts like pottery and ceramics, according to Davis. “This can help identify where these materials were made and how far they were transported. It can also help us to understand the extent of past trading networks.”
Linguistic anthropologists have also used machine intelligence methods to trace the evolution of different languages, Davis said. “Using AI, we can learn when and where languages emerged around the world.”
In other cases, AI has helped reconstruct or decipher ancient texts. Last year, researchers at Google’s DeepMind used a deep neural network called PYTHIA to recreate missing inscriptions in ancient Greek from damaged surfaces of objects made of stone or ceramics.
Named after the Oracle at Delphi, PYTHIA “takes a sequence of damaged text as input, and is trained to predict character sequences comprising hypothesised restorations of ancient Greek inscriptions,” the researchers reported.
In a similar fashion, Chinese scientists applied a convolutional neural network (CNN) to untangle another ancient tongue once found on turtle shells and ox bones. The CNN managed to classify oracle bone morphology in order to piece together fragments of these divination objects, some with inscriptions that represent the earliest evidence of China’s recorded history.
“Differentiating the materials of oracle bones is one of the most basic steps for oracle bone morphology—we need to first make sure we don’t assemble pieces of ox bones with tortoise shells,” lead author of the study, associate professor Shanxiong Chen at China’s Southwest University, told Synced, an online tech publication in China.
AI Helps Archaeologists Get the Scoop…
And then there are applications of AI in archaeology that are simply … interesting. Just last month, researchers published a paper about a machine learning method trained to differentiate between human and canine paleofeces.
The algorithm, dubbed CoproID, compares the gut microbiome DNA found in the ancient material with DNA found in modern feces, enabling it to get the scoop on the origin of the poop.
Also known as coprolites, paleo-feces from humans and dogs are often found in the same archaeological sites. Scientists need to know which is which if they’re trying to understand something like past diets or disease.
“CoproID is the first line of identification in coprolite analysis to confirm that what we’re looking for is actually human, or a dog if we’re interested in dogs,” Maxime Borry, a bioinformatics PhD student at the Max Planck Institute for the Science of Human History, told Vice.
…But Machine Intelligence Is Just Another Tool
There is obviously quite a bit of work that can be automated through AI. But there’s no reason for archaeologists to hit the unemployment line any time soon. There are also plenty of instances where machines can’t yet match humans in identifying objects or patterns. At other times, it’s just faster doing the analysis yourself, Davis noted.
“For ‘big data’ tasks like detecting archaeological materials over a continental scale, AI is useful,” he said. “But for some tasks, it is sometimes more time-consuming to train an entire computer algorithm to complete a task that you can do on your own in an hour.”
Still, there’s no telling what the future will hold for studying the past using artificial intelligence.
“We have already started to see real improvements in the accuracy and reliability of these approaches, but there is a lot more to do,” Davis said. “Hopefully, we start to see these methods being directly applied to a variety of interesting questions around the world, as these methods can produce datasets that would have been impossible a few decades ago.”
Image Credit: James Wheeler from Pixabay Continue reading