Learning Software In A Soft Exosuit

Wearables and robots don’t often intersect, because most robots rely on rigid bodies and programming while we don’t. Exoskeletons are an instance where robots interact with our bodies, and a soft exosuit is even closer to our physiology. Machine learning is closer to our minds than a simple state machine. The combination of machine learning software and a soft exosuit is a match made in heaven for the Harvard Biodesign Lab and Agile Robotics Lab.

Machine learning studies a walker’s steady gait for twenty periods while vitals are monitored to assess how much energy is being expended. After watching, the taught machine assists instead of assessing. This type of personalization has been done in the past, but the addition of machine learning shows that the necessary customization can be programmed into each machine without a team of humans.

Exoskeletons are no stranger to these pages, our 2017 Hackaday Prize gave $1000 to an open-source set of robotic legs and reported on an exoskeleton to keep seniors safe.

Continue reading “Learning Software In A Soft Exosuit”

Hackaday Prize Entry: Two Leg Robot

If you’re working on your own bipedal robot, you don’t have to start from the ground up anymore. [Ted Huntington]’s Two Leg Robot project aims to be an Open Source platform that’ll give any future humanoid-robot builders a leg up.

While we’ve seen quite a few small two-legged walkers, making a pair of legs for something human-sized is a totally different endeavor. [Ted]’s legs are chock-full of sensors, and there’s a lot of software that processes all of the data. That’s full kinematics and sensor info going back and forth from 3D model to hardware. Very cool. And to top it all off, “Two Leg” uses affordable motors and gearing. This is a full-sized bipedal robot platform that you might someday be to afford!

Will walking robots really change the world? Maybe. Will easily available designs for an affordable bipedal platform give hackers of the future a good base to stand on? We hope so! And that’s why this is a great entry for the Hackaday Prize.

Measuring Walking Speed Wirelessly

There are a lot of ways to try to mathematically quantify how healthy a person is. Things like resting pulse rate, blood pressure, and blood oxygenation are all quite simple to measure and can be used to predict various clinical outcomes. However, one you may not have considered is gait velocity, or the speed at which a person walks. It turns out gait velocity is a viable way to predict the onset of a wide variety of conditions, such as congestive heart failure or chronic obtrusive pulmonary disease. It turns out, as people become sick, elderly or infirm, they tend to walk slower – just like the little riflemen in your favourite RTS when their healthbar’s way in the red. But how does one measure this? MIT’s CSAIL has stepped up, with a way to measure walking speed completely wirelessly.

You can read the paper here (PDF). The WiGate device sends out a low-power radio signal, and then measures the reflections to determine a person’s location over time. Alone, however, this is not enough – it’s important to measure the walking speed specifically, to avoid false positives being triggered by a person simply not moving while watching television, for example. Algorithms are used to separate walking activity from the data set, allowing the device to sit in the background, recording walking speed data with no user interaction required whatsoever.

This form of passive monitoring could have great applications in nursing homes, where staff often have a huge number of patients to monitor. It would allow the collection of clinically relevant data without the need for any human intervention; the device could simply alert staff when a patient’s walking pattern is indicative of a bigger problem.

We see some great health research here at Hackaday – like this open source ECG. Video after the break.

Continue reading “Measuring Walking Speed Wirelessly”

Train Your Robot To Walk with a Neural Network

[Basti] was playing around with Artificial Neural Networks (ANNs), and decided that a lot of the “hello world” type programs just weren’t zingy enough to instill his love for the networks in others. So he juiced it up a little bit by applying a reasonably simple ANN to teach a four-legged robot to walk (in German, translated here).

While we think it’s awesome that postal systems the world over have been machine sorting mail based on similar algorithms for years now, watching a squirming quartet of servos come to forward-moving consensus is more viscerally inspiring. Job well done! Check out the video embedded below.

Continue reading “Train Your Robot To Walk with a Neural Network”

Floating Walking Robot

It’s no secret that we love bizarre robot locomotion, so we are naturally suckers for BALLU (YouTube link, also embedded below) the Bouyancy-Assisted Lightweight Legged Unit. The project started with a simple observation — walking robots are constrained by having to hold themselves up — and removing that constraint make success much easier. Instead of walking, BALLU almost floats and uses what little net weight it does have to push against the ground.

Continue reading “Floating Walking Robot”

Vote With Your Feet

Gamifying life is silly, fun, and a great way to interact with those strangers who you pass everyday. Here’s one example that might just pop up along your next walk to work. It’s a way to take a very unscientific straw poll on any topic — you won’t even have to use your hands to cast a ballot.

A group called [Vote With Your Feet] has come up with a novel way of casting ballots. Simply walk down the sidewalk and through one of two doorways, each labeled with either side of a dichotomy. Each doorway is able to count the number of people that pass through it, so any issue imaginable can be polled. They already did vim vs emacs (59 to 27),  and we’d like to see Keynes vs Hayek, or even Ovaltine vs Nesquik. Users can send the machine new issues for the masses to vote on, so the entertainment is quite literally limited only by your imagination.

thumbThe physical build is well documented. Since this is used outside, the choice of a flipdot display (of course always fun to play with) is perfect for this high-contrast in any level of light. Each doorway has a break-beam sensor which is monitored by the Raspberry Pi driving the overhead display (here’s code for it all if you want to dig in).

The point of this art installation like this is to get people to interact with their environment in a novel way, which this project has accomplished exceptionally well. In 3 days, they registered over 10,000 votes which are viewable on their website. If you have a project in mind that calls for data visualization you might want to keep this in your back pocket.

We have also seen other ways that doorways can count people outside of voting, if you’re looking for any inspiration yourself.

Hacklet 83 – Tiny Robot Projects

Hackers, makers, and engineers have been hacking on robot projects since the era of clockwork mechanics. Any robot is a cool project, but there is something particularly attractive about small ones. Maybe it’s the skill required to assemble them, or perhaps it’s the low-cost. Either way, there are lots of palm-sized robot projects on Hackaday.io. This week on the Hacklet, we’re going to highlight a few of them!

tinyrobot2We start with the granddaddy of them all, [shlonkin] and Tiny robot family. [Shlonkin] built line following robots that can hide under a US half-dollar coin. The robots are simple circuits – an ATtiny85 with an LED and pair of phototransistors. The code is provided both in Arduino’s wiring, and in straight C++. Two coreless motors, normally used in cell phones vibrators or quadcopters, provide the locomotion. These robots only know one thing – moving forward and following a line. They do it well though! We love this project so much that we hosted a tiny robot workshop at the 10th anniversary back in 2014.

toteWhen it comes to tiny walking robots, [Radomir Dopieralski] is the king. Many of his projects are small biped, quadruped, or even hexapod robots. He’s done things with 9 gram nano servos that we thought were impossible. Tote, an affordable spider robot, is his latest creation. Tote is a four-legged bot utilizing 12 9 gram servos. [Radomir] created a custom PCB for Tote, which acts as a carrier for its Arduino Pro Mini Brain. This robot is easily expandable – [Radomir] has experimented with the Teensy 3 series as well. Controlling the robot can be anything from an ESP8266 to an infrared remote control.

botbot[Alan Kilian] may well have the ultimate tease project with Hand-wound inductors for a tiny robot. [Alan] was using some tiny GM-10 motors on his micro-bot. The motors didn’t have inductance for the locked-antiphase drive controller. His solution was to wind some coils to provide a bit of added inductance. The mod worked, current consumption dropped from 116 ma to about 6 ma. We want to know more about that ‘bot though! It’s controlled by a Megabitty, [Monty Goodson’s] ATmega8 controller board from sometime around 2003. The lilliputian board has been very popular with the nano sumo crowd. Other than the controller, motors, and the plywood frame, [Alan] has left us guessing about his robot. If you see him, tell [Alan] to give us more info on his micro robot’s design and construction!


espbot[Ccates] jumped on the tiny robot bandwagon with Tiny wi-fi robot. Rather than go with an Arduino for control, [Ccates] grabbed the popular ESP-8266 WiFi module. The construction of the bot is inspired by [shlonkin’s] tiny robot family up above. This bot is controlled by the Xtensa processor embedded in the ESP-8266. Since it only drives forward, it only takes two GPIO pins to control the transistors driving the motors. Even the diminutive ESP-01 module has enough I/O for that. We’d love see some sensors and a full H-bridge on this micro beastie!


If you want to see more palm-sized robot projects, check out our new tiny robot projects list! These ‘bots are small, so I may have missed yours. If that’s the case, don’t be shy, just drop me a message on Hackaday.io. That’s it for this week’s Hacklet. As always, see you next week. Same hack time, same hack channel, bringing you the best of Hackaday.io!