Several video clips of a robot arm manipulating objects in a kitchen environment, demonstrating some of the 12 generalized skills

RoboAgent Gets Its MT-ACT Together

Researchers at Carnegie Mellon University have shared a pre-print paper on generalized robot training within a small “practical data budget.” The team developed a system that breaks movement tasks into 12 “skills” (e.g., pick, place, slide, wipe) that can be combined to create new and complex trajectories within at least somewhat novel scenarios, called MT-ACT: Multi-Task Action Chunking Transformer. The authors write:

Trained merely on 7500 trajectories, we are demonstrating a universal RoboAgent that can exhibit a diverse set of 12 non-trivial manipulation skills (beyond picking/pushing, including articulated object manipulation and object re-orientation) across 38 tasks and can generalize them to 100s of diverse unseen scenarios (involving unseen objects, unseen tasks, and to completely unseen kitchens). RoboAgent can also evolve its capabilities with new experiences.

Continue reading “RoboAgent Gets Its MT-ACT Together”

Measuring Planck’s Constant (Again)

There are many well-known physical constants, but it always interests us when someone can approximately measure them using equipment you probably have. We could pretend it is because we want to help kids do science projects, but who are we really kidding? It is just the cool factor. [Stoppi] shows us several neat ways to measure Planck’s constant (German language, Google Translate link) using things like LEDs, solar cells, and common test equipment. If you don’t want to translate the web page, you can also see the setup and the math behind it in the video below.

If complex math triggers you, this might not be the video for you. The particular test in the video does require a very low current measurement, but that’s not very hard to arrange these days. There are actually several methods covered in the post, and one of them uses one of those familiar “component testers” that has an Atmel CPU, a socket, and an LCD. These can measure the forward current of LEDs, and if you know the wavelength of the LED, you can determine the constant. There’s even a custom device that integrates several LEDs to do the job.

Continue reading “Measuring Planck’s Constant (Again)”

Why Nuclear Bombs Can’t Set The World On Fire

Before the first atomic bomb was detonated, there were some fears that a fission bomb could “ignite the atmosphere.” Yes, if you’ve just watched Oppenheimer, read about the Manhattan Project, or looked into atomic weapons at all, you’ll be familiar with the concept. Physicists determined the risk was “near zero,” proceeded ahead with the Trinity test, and the world lived to see another day.

You might be wondering what this all means. How could the very air around us be set aflame, and how did physicists figure out it wasn’t a problem? Let’s explore the common misunderstandings around this concept, and the physical reactions at play.

Continue reading “Why Nuclear Bombs Can’t Set The World On Fire”

Building A DIY Cloud Chamber

[RCLifeOn] happened to come into possession of some radioactive uranium ore. He thus decided to build a cloud chamber to visualize the products of radioactive decay in a pleasing visual manner.

The construction is fairly straightforward stuff. A 3D-printer build plate was used to heat isopropyl alcohol to a vapor, while a bank of thermoelectric coolers then cool the alcohol down to -30 C to create a dense fog. The build uses a glass chamber with a bank of powerful LEDs to illuminate the fog, making it easier to see the trails from radioactive particles passing through. [RCLifeOn] later used a variety of radioactive sources to deliver a bunch of particles into the chamber for more action, too. He also experimented with blocking particles with a variety of materials.

It’s one of the bigger cloud chambers we’ve seen, and seems to work great. You can build a simple version pretty easily, or you could travel to a local museum or science center if you’re too busy to tackle it at home. Video after the break.

Continue reading “Building A DIY Cloud Chamber”

Giving Solar Power’s Mortal Enemies A Dusting Without Wasting Water

A prerequisite for photovoltaic (PV) and concentrated solar power (CSP) technologies to work efficiently is as direct an exposure to the electromagnetic radiation from the sun as possible. Since dust and similar particulates are excellent at blocking the parts of the EM spectrum that determine their efficiency, keeping the panels and mirrors free from the build-up of dust, lichen, bird droppings and other perks of planetary life is a daily task for solar farm operators. Generally cleaning the panels and mirrors involves having trucks drive around with a large water tank to pressure wash the dirt off, but the use of so much water is problematic in many regions.

Keeping PV panels clean is also a consideration on other planets than Earth. So far multiple Mars rovers and landers have found their demise at the hands of Martian dust after a layer covered their PV panels, and Moon dust (lunar regolith) is little better. Despite repeated suggestions by the peanut gallery to install wipers, blowers or similar dust removal techniques, keeping particulates from sticking to a surface is not as easy an engineering challenge as it may seem, even before considering details such as the scaling issues between a singular robot on Mars versus millions of panels and mirrors on Earth.

There has been research into the use of the electrostatic effect to repel dust, but is there a method that can keep both solar-powered robots on Mars and solar farms on Earth clean and sparkling, rather than soiled and dark?

Continue reading “Giving Solar Power’s Mortal Enemies A Dusting Without Wasting Water”

Building A Peltier-Powered Cloud Chamber

If you’ve been watching Oppenheimer and it’s gotten you all excited about the idea of radioactive decay, you might want to visualize it. A cloud chamber is the perfect way to do that, and [NuclearPhoenix] is here to show us just how to build one.

The build relies on a Peltier device to cool a 10 cm square copper plate down to temperatures as low as -30 °C (-22 °F). Isopropyl alcohol is evaporated via warming resistors within the cloud chamber, and then condenses in the cooled area, creating a thin layer of fog. Ionizing radiation that passes through the chamber can then be spotted by the the trails it leaves through the fog. It’s even possible to identify the type of radiation passing through by the type of trail it leaves. Alpha particles leave shorter traces, while more energetic beta particles which are difficult to stop tend to streak further.

It bears noting that if you see a ton of activity in your cloud chamber at home, it might be worth making some enquiries. Some cloud chambers you’ll see in museums and the like use a small radioactive source to generate some excitement for viewers, though. Video after the break.

Continue reading “Building A Peltier-Powered Cloud Chamber”

Dielectric Mirror Shines Bright

We knew the mirrors in our house were not really very good mirrors, optically speaking. Your mirror eats up 20 to 40 percent of the light that hits it. High-quality first-surface mirrors are better, but [Action Lab] has a video (see below) of something really different: a polymer dielectric mirror with 99.5% reflectivity. In addition, it has no Brewster angle — light that hits it from any angle will reflect.

Turns out something that thin and reflective can be hard to find. It also makes a little flashlight if you roll a tube of the material and pinch the back end together. The light that would have exited the rear of the tube now bounces around until it exits from the front, making it noticeably bright. The film comes from 3M, and apparently, they were surprised about the optical properties, too.

Continue reading “Dielectric Mirror Shines Bright”