Odds are that if you’ve been to the beach or gone camping or somewhere in between, you are familiar with inflatable products like air mattresses. It’s nothing spectacular to see a rectangle inflate into a thicker, more comfortable rectangle, but what if your air mattress inflated into the shape of a crane?
We’ve seen similar ideas in quadcopters and robots using more mechanical means, but this is method uses air instead. To make this possible, the [Tangible Media Group] out of [MIT’s Media Lab] have developed aeroMorph — a program that allows the user to design inflatable constructs from paper, plastic or fabric with careful placement of a few folding joints.
These designs are exported and imprinted onto the medium by a cartesian coordinate robot using a heat-sealing attachment. Different channels allow the medium to fold in multiple directions depending on where the air is flowing, so this is a bit more complicated than, say, a bouncy castle. That, and it’s not often you see paper folding itself. Check it out!
[James Bruton] is well known for making robots using electric motors but he’s decided to try his hand at using pneumatics in order to make a fighting robot. The pneumatic cylinders will be used to give it two powerful punching arms. In true [James Bruton] fashion, he’s started with some experiments first, using the pneumatic cylinders from foot pumps. The cylinders he’s tried so far are taken out of single cylinder foot pumps from Halfords Essentials, costing only £6.29, around $8.11 US. That’s far cheaper than a commercial pneumatic cylinder, and perfectly adequate for this first step.
He did have to hack the cylinder a little though, besides removing it from its mounting and moving it to a DIY frame. Normally when you step down on a foot pump’s lever, you compress the cylinder, forcing air out the hose and into whatever you’re inflating. But he wanted to push air in the other direction, into the hose and into the cylinder. That would make the cylinder expand and thereby extend a robot fighting arm. And preferably that would be done rapidly and forcefully. However, a check valve at the hose outlet prevented air from entering the cylinder from the hose. So he removed the check valve. Now all he needed was a way to forcefully, and rapidly, push air into the hose.
For that he bought a solenoid activated valve on eBay, and a compressor with a 24 liter reservoir and a decent air flow rate of 180 liters per minute. The compressor added £110 ($142) to the cost of his project but that was still cheaper than the batteries he normally buys for his electric motor robots.
After working his usual CAD and 3D printing magic, he came up with an arm for the cylinder and a body that could fit two more valve activated cylinders to act as a working shoulder. A little more 3D printing and electronics, and he had 3 switches, one for each valve and cylinder. He then had the very successful results his experiment. You can see the entire R&D process in the video below, along with demonstrations of the resulting punching robot arm. We think it’s fairly intimidating for a first step.
We’ve always been fascinated at the number of ways logic gates can spring into being. Sure, we think of logic gates carrying electrons, but there are so many other mechanical means to do the same thing. Another method that sometimes has a practical use is fluidic or pneumatic logic. We guess [dAcid] has a similar interest since he’s written two posts on how to construct the gates. One post covers making them with ordinary tools. The other requires an SLA printer.
According to [dAcid], the design is effectively the same either way, but the SLA printing is more precise. Silicone is an important component, either way. Fluidic logic has applications in some mechanical systems, although digital logic has made it less important than it once was. However, it is very possible that nanotechnology systems will implement logic mechanically, so this is still an interesting technique to understand. You can see videos of how a D latch looks using both methods, below.
The I/O capabilities built into most microcontrollers make it easy to measure the analog world. Say you want to build a data logger for temperature. All you need to do is get some kind of sensor that has a linear voltage output that represents the temperature range you need to monitor — zero to five volts representing 0° to 100°C, perhaps. Hook the sensor up to and analog input, whip up a little code, and you’re done. Easy stuff.
Now put a twist on it: you need to mount the sensor far from the microcontroller. The longer your wires, the bigger the voltage drop will be, until eventually your five-volt swing representing a 100° range is more like a one-volt swing. Plus your long sensor leads will act like a nice antenna to pick up all kinds of noise that’ll make digging a usable voltage signal off the line all the harder.
Luckily, industrial process engineers figured out how to deal with these problems a long time ago by using current loops for sensing and control. The most common standard is the 4-mA-to-20-mA current loop, and here we’ll take a look at how it came to be, how it works, and how you can leverage this basic process control technique for your microcontroller projects.
[Bob] wanted to build a pinball-drop-style resetting target that he could use while practicing with his pistol. His first idea was to make the targets sturdy enough for use with 9 mm ammunition, and he planned to use 1/2” thick steel for the targets and 11-gauge steel tubing for the frame. However, the targets weighed 50 pounds together and that was more weight than the pneumatic actuators could lift. He ended up using 1/4” steel and thereby halving weight. The downside was that [Bob] had to switch out the nine for a .22.
Controlling everything is a 555 circuit. When triggered, it opens up a relay for one second, which trips the solenoid valve controlling the pneumatic actuators. Originally he wanted to have switches under each target, and only by dropping all four would the reset circuit be triggered. However, he built a simpler solution: a bulletproof button off to one side–effectively a fifth target–that when triggered resets the targets.
HaD have some pretty good shots in our number but we’d probably end up hitting the pneumatic actuators at least once. [Bob] did add 16-gauge steel sheeting to protect the air lines and wires from bullet splatter, which in his experience is more of a threat than a direct hit.
Whether it’s wheels, tracks, feet, or even a roly-poly body like BB-8, most robots have to deal with an essential problem: dirt and grit can get into the moving bits and cause problems. Some researchers from UCSD have come up with a clever way around this: pneumatically actuated soft-legged robots that adapt to rough terrain.
At a top speed of 20 mm per second, [Michael Tolley]’s squishy little robot won’t set any land speed records. But for applications like search and rescue or placing sensors in inhospitable or inaccessible locations, slow and steady might just win the race. The quadrupedal robot’s running gear can be completely 3D-printed on any commercial printer capable of using a soft filament. The legs each contain three parallel air chambers within a bellowed outer skin; alternating how the chambers are inflated controls how they move. The soft legs adapt to unstructured terrain and are completely sealed, eliminating intrusion problems. The video below shows how the bot gets around just fine over rocks and sand.
The legs remind us a little of our [Joshua Vazquez]’s tentacle mechanism, but with fewer parts. Right now, the soft robot is tethered to its air supply, but the team is working on a miniaturized pump to make the whole thing mobile. At which point we bet it’ll even be able to swim.
If you’ve ever experimented with a robot gripper, you’ll know that while it is easy to make an analogue of the human ability to grip between thumb and forefinger, it is extremely difficult to capture the nuances of grip with the benefit of touch feedback to supply only just enough of the force required to grip and hold an object. You as a human can pick up a delicate eggshell without breaking it using the same hand you might use to pick up a baseball or a cricket ball, but making your robot do the same thing is something of an engineering challenge.
The robot gripper is something that has exercised the minds of the folks at Festo, and the solution they have arrived at is as beautiful as it is novel. They have produced a gripper based upon the action of an octopus tentacle, though unlike the muscle of the real thing they’ve created a silicone tube which bends inwards when inflated. Its inner surface is covered with octopus-like suckers, some of which can be activated by a vacuum. The result is a very capable and versatile gripper which due to its soft construction is ideal for use in environments in which robots and humans interact.
They’ve put up a slick video showing the device in action, which we’ve put below the break. Tasks such as gripping a rolled-up magazine or a plastic bottle that would tax more conventional grippers are performed faultlessly.