Tag Archives: up
#439941 Flexible Monocopter Drone Can Be ...
It turns out that you don't need a lot of hardware to make a flying robot. Flying robots are usually way, way, way over-engineered, with ridiculously over the top components like two whole wings or an obviously ludicrous four separate motors. Maybe that kind of stuff works for people with more funding than they know what to do with, but for anyone trying to keep to a reasonable budget, all it actually takes to make a flying robot is one single airfoil plus an attached fixed-pitch propeller. And if you make that airfoil flexible, you can even fold the entire thing up into a sort of flying robotic swiss roll.
This type of drone is called a monocopter, and the design is very generally based on samara seeds, which are those single-wing seed pods that spin down from maple trees. The ability to spin slows the seeds' descent to the ground, allowing them to spread farther from the tree. It's an inherently stable design, meaning that it'll spin all by itself and do so in a stable and predictable way, which is a nice feature for a drone to have—if everything completely dies, it'll just spin itself gently down to a landing by default.
The monocopter we're looking at here, called F-SAM, comes from the Singapore University of Technology & Design, and we've written about some of their flying robots in the past, including this transformable hovering rotorcraft. F-SAM stands for Foldable Single Actuator Monocopter, and as you might expect, it's a monocopter that can fold up and uses just one single actuator for control.
There may not be a lot going on here hardware-wise, but that's part of the charm of this design. The one actuator gives complete directional control: increasing the throttle increases the RPM of the aircraft, causing it to gain altitude, which is pretty straightforward. Directional control is trickier, but not much trickier, requiring repetitive pulsing of the motor at a point during the aircraft's spin when it's pointed in the direction you want it to go. F-SAM is operating in a motion-capture environment in the video to explore its potential for precision autonomy, but it's not restricted to that environment, and doesn't require external sensing for control.
While F-SAM's control board was custom designed and the wing requires some fabrication, the rest of the parts are cheap and off the shelf. The total weight of F-SAM is just 69g, of which nearly 40% is battery, yielding a flight time of about 16 minutes. If you look closely, you'll also see a teeny little carbon fiber leg of sorts that keeps the prop up above the ground, enabling the ground takeoff behavior without contacting the ground.
You can find the entire F-SAM paper open access here, but we also asked the authors a couple of extra questions.
IEEE Spectrum: It looks like you explored different materials and combinations of materials for the flexible wing structure. Why did you end up with this mix of balsa wood and plastic?
Shane Kyi Hla Win: The wing structure of a monocopter requires rigidity in order to be controllable in flight. Although it is possible for the monocopter to fly with more flexible materials we tested, such as flexible plastic or polymide flex, they allow the wing to twist freely mid-flight making cyclic control effort from the motor less effective. The balsa laminated with plastic provides enough rigidity for an effective control, while allowing folding in a pre-determined triangular fold.
Can F-SAM fly outdoors? What is required to fly it outside of a motion capture environment?
Yes it can fly outdoors. It is passively stable so it does not require a closed-loop control for its flight. The motion capture environment provides its absolute position for station-holding and waypoint flights when indoors. For outdoor flight, an electronic compass provides the relative heading for the basic cyclic control. We are working on a prototype with an integrated GPS for outdoor autonomous flights.
Would you be able to add a camera or other sensors to F-SAM?
A camera can be added (we have done this before), but due to its spinning nature, images captured can come out blurry. 360 cameras are becoming lighter and smaller and we may try putting one on F-SAM or other monocopters we have. Other possible sensors to include are LiDAR sensor or ToF sensor. With LiDAR, the platform has an advantage as it is already spinning at a known RPM. A conventional LiDAR system requires a dedicated actuator to create a spinning motion. As a rotating platform, F-SAM already possesses the natural spinning dynamics, hence making LiDAR integration lightweight and more efficient.
Your paper says that “in the future, we may look into possible launching of F-SAM directly from the container, without the need for human intervention.” Can you describe how this would happen?
Currently, F-SAM can be folded into a compact form and stored inside a container. However, it still requires a human to unfold it and either hand-launch it or put it on the floor to fly off. In the future, we envision that F-SAM is put inside a container which has the mechanism (such as pressured gas) to catapult the folded unit into the air, which can begin unfolding immediately due to elastic materials used. The motor can initiate the spin which allows the wing to straighten out due to centrifugal forces.
Do you think F-SAM would make a good consumer drone?
F-SAM could be a good toy but it may not be a good alternative to quadcopters if the objective is conventional aerial photography or videography. However, it can be a good contender for single-use GPS-guided reconnaissance missions. As it uses only one actuator for its flight, it can be made relatively cheaply. It is also very silent during its flight and easily camouflaged once landed. Various lightweight sensors can be integrated onto the platform for different types of missions, such as climate monitoring. F-SAM units can be deployed from the air, as they can also autorotate on their way down, while also flying at certain periods for extended meteorological data collection in the air.
What are you working on next?
We have a few exciting projects on hand, most of which focus on 'do more with less' theme. This means our projects aim to achieve multiple missions and flight modes while using as few actuators as possible. Like F-SAM which uses only one actuator to achieve controllable flight, another project we are working on is the fully autorotating version, named Samara Autorotating Wing (SAW). This platform, published earlier this year in IEEE Transactions on Robotics , is able to achieve two flight modes (autorotation and diving) with just one actuator. It is ideal for deploying single-use sensors to remote locations. For example, we can use the platform to deploy sensors for forest monitoring or wildfire alert system. The sensors can land on tree canopies, and once landed the wing provides the necessary area for capturing solar energy for persistent operation over several years. Another interesting scenario is using the autorotating platform to guide the radiosondes back to the collection point once its journey upwards is completed. Currently, many radiosondes are sent up with hydrogen balloons from weather stations all across the world (more than 20,000 annually from Australia alone) and once the balloon reaches a high altitude and bursts, the sensors drop back onto the earth and no effort is spent to retrieve these sensors. By guiding these sensors back to a collection point, millions of dollars can be saved every year—and also [it helps] save the environment by polluting less. Continue reading
#439902 Boston Dynamics robots imitate Rolling ...
The team at robotics company Boston Dynamics has released a video promoting itself while also honoring the Rolling Stones—this year marks the 40th anniversary of the release of the song 'Start Me Up.' The release of the song was notable also for the video that accompanied the song, with the members of the group playing their instruments and lead singer Mick Jagger strutting around on stage. Continue reading
#439110 Robotic Exoskeletons Could One Day Walk ...
Engineers, using artificial intelligence and wearable cameras, now aim to help robotic exoskeletons walk by themselves.
Increasingly, researchers around the world are developing lower-body exoskeletons to help people walk. These are essentially walking robots users can strap to their legs to help them move.
One problem with such exoskeletons: They often depend on manual controls to switch from one mode of locomotion to another, such as from sitting to standing, or standing to walking, or walking on the ground to walking up or down stairs. Relying on joysticks or smartphone apps every time you want to switch the way you want to move can prove awkward and mentally taxing, says Brokoslaw Laschowski, a robotics researcher at the University of Waterloo in Canada.
Scientists are working on automated ways to help exoskeletons recognize when to switch locomotion modes — for instance, using sensors attached to legs that can detect bioelectric signals sent from your brain to your muscles telling them to move. However, this approach comes with a number of challenges, such as how how skin conductivity can change as a person’s skin gets sweatier or dries off.
Now several research groups are experimenting with a new approach: fitting exoskeleton users with wearable cameras to provide the machines with vision data that will let them operate autonomously. Artificial intelligence (AI) software can analyze this data to recognize stairs, doors, and other features of the surrounding environment and calculate how best to respond.
Laschowski leads the ExoNet project, the first open-source database of high-resolution wearable camera images of human locomotion scenarios. It holds more than 5.6 million images of indoor and outdoor real-world walking environments. The team used this data to train deep-learning algorithms; their convolutional neural networks can already automatically recognize different walking environments with 73 percent accuracy “despite the large variance in different surfaces and objects sensed by the wearable camera,” Laschowski notes.
According to Laschowski, a potential limitation of their work their reliance on conventional 2-D images, whereas depth cameras could also capture potentially useful distance data. He and his collaborators ultimately chose not to rely on depth cameras for a number of reasons, including the fact that the accuracy of depth measurements typically degrades in outdoor lighting and with increasing distance, he says.
In similar work, researchers in North Carolina had volunteers with cameras either mounted on their eyeglasses or strapped onto their knees walk through a variety of indoor and outdoor settings to capture the kind of image data exoskeletons might use to see the world around them. The aim? “To automate motion,” says Edgar Lobaton an electrical engineering researcher at North Carolina State University. He says they are focusing on how AI software might reduce uncertainty due to factors such as motion blur or overexposed images “to ensure safe operation. We want to ensure that we can really rely on the vision and AI portion before integrating it into the hardware.”
In the future, Laschowski and his colleagues will focus on improving the accuracy of their environmental analysis software with low computational and memory storage requirements, which are important for onboard, real-time operations on robotic exoskeletons. Lobaton and his team also seek to account for uncertainty introduced into their visual systems by movements .
Ultimately, the ExoNet researchers want to explore how AI software can transmit commands to exoskeletons so they can perform tasks such as climbing stairs or avoiding obstacles based on a system’s analysis of a user's current movements and the upcoming terrain. With autonomous cars as inspiration, they are seeking to develop autonomous exoskeletons that can handle the walking task without human input, Laschowski says.
However, Laschowski adds, “User safety is of the utmost importance, especially considering that we're working with individuals with mobility impairments,” resulting perhaps from advanced age or physical disabilities.
“The exoskeleton user will always have the ability to override the system should the classification algorithm or controller make a wrong decision.” 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