AI News, UC Berkeley's Little Legged Robots Grow Wings and Tails

UC Berkeley's Little Legged Robots Grow Wings and Tails

When researchers at UC Berkeley figured out what kinds of awesome things were possible when you endowed a robot with an actuated tail that it could use for pitch control, it earned them an article inNature, which is only slightly less prestigious than an article inIEEE Spectrum.

Anyway, the Berkeley students have been exploring what else you can do when you give ground robots bio-inspired accessories, and they've got some little legged robots doing cool new stuff thanks to the addition of wings and tails.

By combining yaw and pitch to move the tail in a sort of circular motion, the robot can correct pitch, yaw, and roll, meaning that you can fling it however you want and it’ll be able to land on its feet.

The peak controllable turning rate of TAYLRoACH is about 360 degrees per second with an accuracy of close to five degrees, even while the robot is running along at over three body lengths per second.

Because the method of turning is completely independent of the legs, turning the robot doesn't slow it down, but more importantly, it doesn't require a lot of gripping force, so the robot can turn even on very slippery surfaces.

The tail can be slowly moved back to zero and turn the robot again, but a more intriguing idea presented in the paper is to just 'develop a tail with an unlimited range of motion.'

When you get little tiny robots to run around super fast on legs, rotational stability starts to become a problem, but a pair of wings works wonders for allowing the robot to traverse obstacles at high speed without flipping itself over:

Salto-1P Is the Most Amazing Jumping Robot We've Ever Seen

Last December, Duncan Haldane (whose research on incredibly agile bioinspired robots we’ve featured extensively in the past) ended up on the cover of the inaugural issue of Science Robotics with his jumping robot, Salto.

Haldane mentioned to us in December that future work on Salto could include chaining together multiple jumps, and in a paper just accepted to the 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), he and co-author Justin Yim at UC Berkeley’s Biomimetic Millisystems Lab, led by ProfessorRonald Fearing,show the improvements that they’ve made over the last six months.

The galago is able to manage this thanks to a rather clever bit of leg design which uses variable mechanical advantage, leveraging the shape of their leg to amplify the force that their muscles can deliver.

Salto-1P is, according toHaldane, essentially “Salto with half of a mini-quadrotor glued to it.” Those two little thrusters are able to control Salto-1P’s yaw and roll: When they’re thrusting in different directions, the robot yaws, and when they both thrust in the same direction, the robot rolls.

Other hardware modifications include a deeper crouch than the original Salto, which allows more energy to be transferred from the jumping motor into the spring, giving it the highest vertical jumping agility of any battery powered robot at 1.83 m/s.

Haldane says one issue that came when they redesignedthe leg mechanism to allow the robot tojump higher is that, as he puts it,“Salto lost its friendly and forgiving nature.” The robot would occasionally “fire pieces of itself across the room when the motor tore the leg-mechanism apart.” They had to do revise the designto keep everything in one piece.

In fact, 92 percent of the time, the robot is in the air, which means that you really have to control it in the air, which is why the tail and thrusters are necessary (as opposed to control through the leg and foot).

This results in enormous accelerations (on the order of 14 g’s), and to put this in context, Haldane compares Salto-1P to a cheetah: The robot has“a lower duty cycle than a single cheetah limb at top speed,” he says, adding: “Imagine a cheetah running at top speed using only one leg, and then cut the amount of time that leg spends on the ground in half.

That’s the duty factor of Salto-1P.” It’s important to note that when you see Salto-1P bouncing around in the video, it’s doing so untethered, but not completely autonomously: There’s a bunch of stuff going on in the background to get it to perform the way it does.

The path it follows relies on motion capture, with an offboard computer (though not a particularly powerful one) receiving tracking data and wirelessly sending control commands to the robot.

“It’s also useful for gathering performance data since we can very closely track Salto-1P throughout its hopping.” It’s also worth noting that Salto-1P isn’t doing a lot of sensing on its own, and it’s still able to handle all those obstacles at the end of the video, which is impressive.

We brainstormed about 40 different ways to control Salto-1P’s orientation including multiple tails, a single multi-degree-of-freedom inertial tail (like the body of the original Raibert hopper and Disney’s LEAP robot), and big steerable wings.

large part of the engineering effort in designing Salto’s leg was making the complex system of linkages and gears behave like a simple spring moving up and down in a straight line with a variable mechanical advantage.

<?xml version="1.0" encoding="UTF-8"?>Rapid Inversion: Running Animals and Robots Swing like a Pendulum under Ledges

To test the hypothesis that cockroaches and geckos configure their legs and bodies during rapid inversion to swing like a pendulum, we applied a template based on pendulum dynamics that has been used to understand cyclic animal movements such as walking, running and brachiating.

Even using the simplest possible model for a single swing, a point mass with a massless support arm, both brachiation gaits display substantial pendular exchange between kinetic and potential energy [32].

Applying a physical pendulum model with zero transfer of kinetic energy to rapid inversion revealed substantial deviations from the actual trajectories in both the cockroaches and geckos, especially at the onset of the swing (Fig.

The same pendulum with nonzero initial kinetic energy better represented the beginning portion of the swing cycle of both animal trajectories suggesting that energy is effectively transferred from running to swinging, but not without losses (Fig.

In the case of rapid inversion, it is also likely that energy losses occur due to a discontinuity in trajectory, particularly during the transition from running to swinging which requires a redirection of the available kinetic energy.

Given that rapid inversion is not a repeated, sustained activity like brachiation, we hypothesize that energy saving is less important, whereas effective transfer of energy to complete the behavior as quickly as possible with a sufficient level of stability is paramount.