We all know how important it is to achieve balance in life, or at least so the self-help industry tells us. How exactly to achieve balance is generally left as an exercise to the individual, however, with varying results. But what about our machines? Will there come a day when artificial intelligences and their robotic bodies become so stressed that they too will search for an elusive and ill-defined sense of balance?

We kid, but only a little; who knows what the future field of machine psychology will discover? Until then, this kinetic sculpture that achieves literal balance might hold lessons for human and machine alike. Dubbed In Medio Stat Virtus, or “In the middle stands virtue,” [Astrid Kraniger]’s kinetic sculpture explores how a simple system can find a stable equilibrium with machine learning. The task seems easy: keep a ball centered on a track suspended by two cables. The length of the cables is varied by stepper motors, while the position of the ball is detected by the difference in weight between the two cables using load cells scavenged from luggage scales. The motors raise and lower each side to even out the forces on each, eventually achieving balance.

The twist here is that rather than a simple PID loop or another control algorithm, [Astrid] chose to apply machine learning to the problem using the Q-Behave library. The system detects when the difference between the two weights is decreasing and “rewards” the algorithm so that it learns what is required of it. The result is a system that gently settles into equilibrium. Check out the video below; it’s strangely soothing.

We’ve seen self-balancing systems before, from ball-balancing Stewart platforms to Segway-like two-wheel balancers. One wonders if machine learning could be applied to these systems as well.

For many of us. the holiday season is coming up and that means hosting parties and mixing drinks, which can get tiresome. [GreatScott] has come up with a solution, what he calls a crude cocktail mixing machine. But don’t be fooled — it may look crude on the surface, and vibrate a bit while working, but the mechanism is plenty sound and functional.

The machine can mix three different liquids and does so using three peristaltic pumps. In typical [GreatScott] style, while he tears apart the pumps to replace the tubes, he gives us a good glimpse of just how they work. Using a knob and LCD screen, you can enter any quantity you want for the three liquids, though you’ll have to edit the Arduino code if you want to change the liquids’ names.

How does the machine know when to stop pumping a certain liquid? Each pump is rated for a specific quantity per second, though he tests this for each liquid anyway and finds a slight variation which he accounts for in the code. After the machine turns a pump on, a load cell located under the glass tells it when liquid has started arriving at the glass. A simple calculation based on the pump’s quantity per second and the desired quantity tells it how long to leave the pump on for. When the times up, it stops the pump. The result is a machine that’s sure to be a centerpiece for any hacker-filled party. Check out his build and the pump in action in the video below.

But parties need more than just drinks, they also need cookies. So to that end, check out [Ben Krasnow]’s equally cool cookie making machine.

Ever obsessed with stripping the hype from the reality of power tool marketing, and doing so on the cheap, [arduinoversusevil] has come up with a home-brew digital torque meter that does the job of commercial units costing hundreds of times as much.

For those of us used to [AvE]’s YouTube persona, his Instructables post can be a little confusing. No blue smoke is released, nothing is skookum or chowdered, and the weaknesses of specific brands of tools are not hilariously enumerated. For that treatment of this project, you’ll want to see the video after the break. Either way you choose, he shows us how a $6 load cell and a $10 amplifier can be used to accurately measure the torque of your favorite power driver with an Arduino. We’ve seen a few projects based on load cells, like this posture-correcting system, but most of them use the load cell to measure linear forces. [AvE]’s insight that a load cell doesn’t care whether it’s stretched or twisted is the key to making a torque meter that mere mortals can afford.

Looks like low-end load cells might not be up to measuring the output on your high-power pneumatic tools, at least not repeatedly, but they ought to hold up to most electric drivers just fine. And spoiler alert: the Milwaukee driver that [AvE] tested actually lived up to the marketing.

How can you not be interested in a project that uses load cells, Bluetooth, a Raspberry Pi, and Twitter. Even for those of our readers without a cat, [Scott's] tweeting litter box is worth the read.

Each aspect of this project can be re-purposed for almost any application. The inexpensive load cells, which available from eBay and other retailers, is used to sense when a cat is inside the litter box. Typically sensors like the load cell (that contain a strain gauge) this use a Wheatstone bridge, which is very important for maximizing the sensitivity of resistive sensor. The output then goes to a HX711, which is an ADC specifically built for load cells. A simple alternative would be using an instrumentation amplifier and the built-in ADC of the Arduino. Now, the magic happens. The weight reading is transmitted via an HC-06 Bluetooth module to a Raspberry Pi. Using a simple Perl script, the excreted weight, duration, and the cat’s resulting body weight is then tweeted!

Very nice work! This is a well thought out project that we could see being expanded to recognize the difference between multiple cats (or any other animal that goes inside).