Google Nest Mini Gutted And Rebuilt To Run Custom Agents

July 23, 2023 by Zoe Skyforest 19 Comments

The Google Nest Mini is a popular smart speaker, but it’s very much a cloud-based Big Tech solution. For those that want to roll their own voice assistant, or just get avoid the corporate surveillance of it all, [Justin Alvey’s] work may appeal. (Nitter)

[Justin] pulled apart a Nest Mini, ripped out the original PCB, and kitted it out with his own internals. He uses the ESP32 as the basis of his design, since it provides plenty of processing power and WiFi connectivity. His replacement PCB also interfaces with the LEDs, mute switch, and capacitive touch features of the Nest Mini, for ease of interaction.

As a demo, he set up the system to work with a custom “Maubot” assistant using the Matrix framework. He hooked it up with Beeper, a messaging client that collates all your other messaging platforms into one easily-accessible place. The assistant employs GPT3.5, prompted with a list of his family, friends, and other details, to enable him to make calls, send messages, and handle natural language queries. The demo itself is very impressive, and we’d love to try setting up a similar assistant ourselves. Seeing two of [Justin’s] builds talking to each other is amusing, too.

If you’re more comfortable working with Google Assistant rather than dropping it entirely, we’ve looked at that kind of thing, too. Video after the break.

Continue reading “Google Nest Mini Gutted And Rebuilt To Run Custom Agents” →

RoboPianist Is A Simulation For Advancing Robotic Control

July 21, 2023 by Zoe Skyforest 10 Comments

Researchers at Google have posed themselves an interesting problem to solve: mastering the piano. However, they’re not trying to teach themselves, but a pair of simulated anthropomorphic robotic hands instead. Enter RoboPianist.

The hope is that the RoboPianist platform can help benchmark “high-dimensional control, targeted at testing high spatial and temporal precision, coordination, and planning, all with an underactuated system frequently making-and-breaking contacts.”

If that all sounds like a bit much to follow, the basic gist is that playing the piano takes a ton of coordination and control. Doing it in a musical way requires both high speed and perfect timing, further upping the challenge. The team hopes that by developing control strategies that can master the piano, they will more broadly learn about techniques useful for two-handed, multi-fingered control. To that end, RoboPianist models a pair of robot hands with 22 actuators each, or 44 in total. Much like human hands, the robot hands are underactuated by design, meaning they have less actuators than their total degrees of freedom.

Continue reading “RoboPianist Is A Simulation For Advancing Robotic Control” →

AI Learns To Walk In 3D Training Grounds

July 21, 2023 by Zoe Skyforest 13 Comments

AI agents are learning to do all kinds of interesting jobs, even the creative ones that we quite prefer handling ourselves. Nevertheless, technology marches on. Working in this area is YouTuber [AI Warehouse], who has been teaching an AI to walk in a simulated environment.

Albert needed some specific guidance to learn how to walk upright, something that humans tend to figure out innately.

The AI controls a vaguely humanoid-like creature, albeit with a heavily-simplified body and limbs. It “lives” in a 3D environment created in the Unity engine, which provides the necessary physics engine for the work. Meanwhile, the ML-Agents package is used to provide the brain for Albert, the AI charged with learning to walk.

The video steps through a variety of “deep reinforcement learning” tasks. In these, the AI is rewarded for completing goals which are designed to teach it how to walk. Albert is given control of his limbs, and simply charged with reaching a button some distance away on the floor. After many trials, he learns to do the worm, and achieves his goal.

Getting Albert to walk upright took altogether more training. Lumpy ground and walls in between him and his goal were used to up the challenge, as well as encouragements to alternate his use of each foot and to maintain an upright attitude. Over time, he was able to progress through skipping and to something approximating a proper walk cycle.

One may argue that the teaching method required a lot of specific guidance, but it’s still a neat feat to achieve nonetheless. It’s altogether more complex than learning to play Trackmania, we’d say, and that was impressive enough in itself. Video after the break.

Continue reading “AI Learns To Walk In 3D Training Grounds” →

Sand Drawing Plotter Runs On ESP32

July 21, 2023 by Zoe Skyforest 3 Comments

Humans have always drawn lines in the sand, whether it’s to communicate a plan of attack or to indicate metaphorically a very real boundary. It’s also something we do just for the aesthetic pleasure, and this plotter from [aidenvigue] is great at performing in just that role.

The plotter traces patterns in the circular sand tray by dragging a small marble with a magnet. This is achieved with a pair of NEMA 17 stepper motors, set up in a polar coordinate fashion. One stepper motor controls the angle, while another motor controls the marble’s distance from the center point of the circle. It’s a simple way to build a circular plotter, and works far better than a Cartesian setup would for this geometry. The build uses an ESP32 as the brains of the operation. It hosts a web interface that allows various patterns to be selected and run on the device. It also runs a set of addressable SK6812 LEDs that light the sand rather nicely.

We’ve seen some great sand plotters before, and have always been particular fans of the larger variety. Video after the break.

Continue reading “Sand Drawing Plotter Runs On ESP32” →

Turning A Quartz Clock Module Into A Time Reference

July 21, 2023 by Zoe Skyforest 11 Comments

If you’re looking for a 1-second time reference, you’d probably just grab a GPS module off the shelf and use the 1PPS output. As demonstrated by [InazumaDenki], though, an old quartz clock module can also do the job with just a little work.

The module was harvested from an old Seiko wall clock, and features the familiar 32.768 KHz crystal you’d expect. This frequency readily divides down by 2 multiple times until you get a useful 1 Hz output. The module, originally designed to run a clock movement, can be repurposed with some basic analog electronics to output a useful time reference. [InazumaDenki] explains various ways this can be done, before demonstrating his favored method by building the device and demonstrating it with a decade counter.

It has some benefits over a GPS time reference, such as running at a much lower voltage and needing no external signal inputs. However, it’s also not going to be quite as accurate. Whether that matters to you or not depends on your specific application. Video after the break.

Continue reading “Turning A Quartz Clock Module Into A Time Reference” →

An Easy Z80 And VGA Upgrade For The Apple II

July 20, 2023 by Zoe Skyforest 15 Comments

The Apple II was at the forefront of the home computer revolution when it came out in 1977. In its era, nobody really cared about hooking up the Apple II to a VGA monitor, but these days, it’s far easier than sourcing an original monitor. The V2 Analog is a useful tool that will let you do just that, plus some other neat tricks, besides.

As demonstrated on Youtube by [Adrian’s Digital Basement], The V2 Analog is basically a slot-in video card for the Apple II, II+, and IIe. It’s based upon the AppleII-VGA, which uses a Raspberry Pi Pico to snoop the 6502 CPU bus and copy the video memory. It then outputs a high-quality VGA signal that is far nicer than the usual composite output options.

As a bonus, the V2 Analog can be reconfigured to run as an emulated AppliCard Z80 expansion card instead. This card was originally intended to allow Apple II users to run CP/M applications. The V2 Analog does a great job in this role, though it bears noting it can’t handle VGA output and Z80 emulation at the same time.

Project files are available on Github for the curious. The Apple II may be long out of production, but it’s certainly not forgotten. Video after the break.

Continue reading “An Easy Z80 And VGA Upgrade For The Apple II” →

Giant 3D Printer Can Print Life-Sized Human Statues

July 20, 2023 by Zoe Skyforest 30 Comments

We’ve seen a few makers 3D scan themselves, and use those to print their own action figures or statuettes. Some have gone so far as building life-sized statues composed of many 3D printed parts. [Ivan Miranda] is no regular maker though, and his custom 3D printer is big enough that he can print himself a life-sized statue in one go.

The printer is a gargantuan thing, using an aluminium frame and a familiar Cartesian layout. It boasts a build volume of 1110 mm x 1110 mm x 2005 mm, making it more than big enough to print human-sized statues. Dogs, cats, and some great apes may be possible, too.

Many of the components are 3D printed, including the various braces and adapters that hold the frame together. The build uses NEMA 23 stepper motors, with Duet3D hardware running the show. Notably, it uses V-wheels for the Z-axis, as linear rails would be prohibitively expensive at the sizes required.

[Ivan] shows off the printer by having it produce a statue of his body at 1:1 scale. It’s not a perfect print, with some layer shifts and an awkward moments where the filament supply was interrupted. It took 108 hours in total, with 76 hours of that being actual print time, and is made up of 4375 layers. Despite its flaws, its an incredibly impressive way to demonstrate the capabilities of the machine.

Eager to build such a printer for yourself? [Ivan] will sell you the design files for a reasonable fee.

[Ivan]’s giant printer was once a large tabletop affair; just look how far it’s come. He’s even come up with a system for using smaller printers to create large-scale construction kits, too. We can’t wait to see what mad project he comes up with next. Video after the break.

Continue reading “Giant 3D Printer Can Print Life-Sized Human Statues” →