Schlagwort: raspberrypi

Reading Time: 3 minutes

Joseph Uding’s robot-building education started with Stan Winston’s Hollywood animatronics and MIT’s robotics online courses. He’s now a project officer at Hive Colab’s department of Robotics and AI.

The single-board computer features prominently in Joseph’s Project Zed robot, created as a means of encouraging citizens to take sensible health precautions. “A friend said ‘build a robot that can help in fighting Ebola in West Africa’. That was in 2016 and we didn’t take it seriously because no cases were in Uganda, but when 2020 came I got to work. With a robot, he says, “you can carry out random tests and collect information from someone without risking important medical workers.”

Raspberry Pi enabled Joseph to add personality and voice-based interaction to the robots he’s built, as well as delivering preset messages such as “don’t forget to wear a face mask”.

Parts and labour

Project Zed is, by necessity, a masterclass in upcycling: 3D printers and next-day component deliveries aren’t reliably available in Uganda, but Joseph believes that if you can have a clear idea of what you want to make, you can make it from almost anything. “The first thing is to see the problem you want to solve with the robot you want to make.” His project took him just three weeks.

After work one day, he went to visit some guys who fix printers and asked them for old dead ones. “They smiled and gave me about four spoiled old printers. From there I got DC motors and plastic parts to use when housing the robot.” The motors would be used to replace those in a broken RC car that would act as the robot’s base and help stabilise it.

The initial designs were in lightweight wood he sourced himself – “just like the way planes are made” – which was light and easy to work with. Plastic from the old printers was used for the casing.

Aside from the Raspberry Pi to control things, the half-humanoid, half-RC car robot has a Roboclaw 2 motor controller and a Google Coral AI USB accelerator .

It can be controlled from a distance of up to roughly 400 metres and can listen, talk, and see objects. “It can move its arms up and down, but not the fingers yet,” says Joseph. “It’s limited to doing one thing at once – bending down to pick something up or moving itself along – because it has only one Raspberry Pi for the head where the AI is based.”

Software trials

Joseph first tried using the open-source TensorFlow and YOLO for object detection, but really had in mind something more like a Movidius Neural Compute Stick to handle face detection (which he didn’t have). Instead, having consulted projects such as the Personal Assistant Robot, he turned to SSH and live video streaming, which did a great job. “I can control the robot from anywhere around the world using my cell phone or PC any time I want to,” Joseph explains – something with clear potential in war or disaster zones where medical workers’ lives would otherwise be at risk.

“Doing such projects where you don’t have the actual resources helps you to be good at recycling things and also teaches you how to be super-creative in life,” he says, modestly.

Reading Time: 2 minutes

Since then he’s made several Raspberry Pi projects, including Spider-Man specs – hence his profile pic. His Mind Patterning headset is intended “to kick off a much larger project using mind-reading capabilities.”

The project reads brainwave data with a MUSE headband and uses TensorFlow machine learning to train a model on the data to recognise a relaxed or non-relaxed mind state.

Mind over matter

Michael used a Raspberry Pi 4 for the Mind Patterning project “because it allows lots of applications to run simultaneously.” The Bluetooth setup required for Raspberry Pi to import data from the headset turned out to be one of the trickier aspects of the build since the latter is really designed to work with iOS or Android.

“Using Raspberry Pi was essential for the TensorFlow tech required for training a neural network on the recorded EEG data,” he explains. “The code is where the design and development was [see his GitHub repo]. I used some libraries from elsewhere in order to get the EEG data logged and accessed. So many people use Raspberry Pi and it ensures that there is always support and information out there to learn from and assist,” he notes.

Raspberry Pi 4’s flexibility and processing power meant Michael needed little else to realise his project.“It has so much hardware on board, including the power for the machine learning, which is where the multicore and lots of RAM really helped,” he says.

Trial and error

“Initial testing of the model training and inference was done with the PipelineRunnerTest.py file,” explains Michael. “This creates artificial sets of training and neural network testing data. It calls CSV preprocessing (which loads and trains the model), then runs that against the test data.”

This approach allowed him to see quickly how well the model would work if given real-world data, and to make adjustments. “I was trying to get the raw EEG sensor data at first, but then realised I needed to get the data processed into alpha/beta/theta wave data before passing to the neural net.”

Michael hopes to develop the project to read more mind states, possibly even emotions, and sees future potential for the mind-patterning idea in simulating human responses via a neural network.

“With more tweaking and more data, it could be used to interact with the hardware without having to manually press buttons, but I’m sure Elon Musk’s Neuralink is going to be far, far more effective than anything I could ever do,” he says.

Reading Time: 2 minutes

pi3g plans to celebrate AIY Projects and Coral by giving away five AIY Voice Lot v2, AIY Vision, or Coral USB Accelerator units to The MagPi readers. And we will make sure the world gets to see your creation!

All you have to do is tell us what you plan to make with your AIY Voice Kit, AIY Vision Kit, or USB Accelerator. If your project is selected, pi3g will send you the kit you need and you will appear in a future edition of The MagPi magazine.

The AIY Voice Kit v2 enables you to explore voice recognition and natural language.

Using AIY Vision Kit you build an intelligent camera that identifies objects.

Both AIY Voice v2 and AIY Vision kits include a Raspberry Pi Zero WH and the cardboard DIY cases (for easy assembly and no soldering required!).

The Coral USB Accelerator brings the Edge TPU (Tensor Processing Unit) to Raspberry Pi. TPU is Google’s purpose-built chip designed to run AI at the edge: your projects are capable of performing fast and efficient AI calculations. Add features such as real-time object recognition and classification, speech processing, and smart decision making. All while maintaining data privacy! It goes great with Raspberry Pi 4’s USB 3.0 port.

So don’t delay! Let us know today what you could do with artificial intelligence.

Reading Time: 7 minutes

Copyright alert!

DOSBox is an emulator and we use it with open-source FreeDOS code. Be mindful of copyright when downloading image files, and only use proprietary software that you own and in accordance with the licence terms.

To run Windows 98 software on Raspberry Pi you’ll need
:

Find a copy of Windows 98

Getting hold of Windows 98 is no longer easy as – unlike MS-DOS, or Windows 3.1 and 7 – Microsoft is no longer allowed to distribute Windows 95, 98, and Me through its Visual Studio Subscriptions program due to an intellectual property claim by Sun.

That means you’ll have to find a second-hand or unsold stock copy – eBay is a good bet for this – or rummage through your loft for old installation discs. The Windows product key is the really important bit here as the software is useless without it: check for key stickers on your old laptop or desktop PCs.

You’ll also need a boot floppy to run most versions of the installation disc.

Image install media

It’s easiest to install Windows 98 from images, so we’ll want to copy both boot and install media. You can use another computer to image your discs and then copy them over, but if you have USB CD and floppy disk drives that play nicely with Raspberry Pi – not all do – then you can use dd in a Terminal window.

CD example:

dd if=/dev/sr0 of=win98.iso

Floppy example:

dd bs=512 count=2880 if=/dev/sda of=win98boot.img

In this tutorial, we’ve put all our media images in our home directory, but you may wish to create dedicated CD and floppy directories to house them in.

Install DOSBox-X

Open a Terminal window and enter:

sudo apt install automake libncurses-dev nasm libsdl-net1.2-dev libpcap-dev libfluidsynth-dev ffmpeg libavdevice58 libavformat-* libswscale-* libavcodec-*
git clone https://github.com/joncampbell123/dosbox-x.git
cd dosbox-x
./build
sudo make install
dosbox-x

DOSBox-X should open at its Z: prompt. Now exit, because we’ll be using a custom config file for our Windows 98 shenanigans.

Before we get started in earnest, set Raspberry Pi’s desktop resolution to 1280×720 under Preferences > Screen Configuration, accessible from the main menu: this improves full-screen performance under Windows 98.

Configure networking

Download our win98_install.conf and win98.conf files from The MagPi GitHub and put them in your home directory.

[sdl]
autolock=true
fullresolution = desktop
output = opengl [dosbox]
title=Windows 98
memsize=64
cycles=60000
vesa modelist width limit=0
vesa modelist height limit=0 [render]
# aspect ratio correction
aspect = true [cpu]
cputype=pentium_mmx
core=normal [sblaster]
sbtype=sb16vibra
irq=5 [ne2000]
# If you want networking in Windows, set ne2000=true.
# This also requires that you set realnic= to a suitable value for your PC - that should be 1 for Raspberry Pi 4 wlan
ne2000=true
nicirq=10
realnic=1 [fdc, primary]
int13fakev86io=true [ide, primary]
int13fakeio=true
int13fakev86io=true [ide, secondary]
int13fakeio=true
int13fakev86io=true
cd-rom insertion delay=4000 [render]
scaler=none [autoexec]
imgmount 2 win98hd.img -size 512,63,130,1023 -fs none
imgmount d win98.iso
imgmount a win98boot.img -t floppy
# If you have a bootable Win98 disc replace the above line with:
# imgmount 0 -el-torito D -t floppy -fs none
boot A:

These files enable NE2000 networking, with NIC 1 selected. This is correct if you’re using Raspberry Pi 4’s integrated wireless LAN. However, if you encounter any issues or wish to use a different internet connection, set reallnic=list in the DOSBox-X config file and check the Network Interfaces List that DOSBox-X will show in the Bash terminal window.

You’ll need the PCAP library to give DOSBox-X network access.

sudo apt install pcap*
sudo setcap cap_net_raw+ep /usr/bin/dosbox-x

Check that this worked:

getcap /usr/bin/dosbox-x

Note that you’ll have to rerun the setcap command whenever you upgrade DOSBox-X.

Create a Windows 98 disk image

From a Terminal, run:

dosbox-x -conf win98_install.conf

Now, from inside DOSBox-X:

imgmake win98hd.img -t hd_2gig -nofs

We’re creating a 2GB hard disk image here, as 4GB images (although supported on other systems) currently throw up virtual disk geometry errors on Raspberry Pi.
We’ve included, and partially commented out, lines to automatically mount and boot the images we’ll use for installation in our win98_install.conf file, so once you’ve created this image, you can either uncomment them and restart DOSBox-X or enter them manually from at the DOS command prompt, as detailed in the following step.

[sdl]
# set fullscreen true if you want to launch in fullscreen mode
fullscreen=false
autolock=true
fullresolution = desktop
output = opengl [dosbox]
title=Windows 98
memsize=512
cycles=60000
vesa modelist width limit=0
vesa modelist height limit=0 [render]
# aspect ratio correction
aspect=true [cpu]
# pentium_mmx is recommended for Windows 98 installs but causes audio distortion in some games: switch cputype to auto if this occurs
cputype=pentium_mmx
core=normal [sblaster]
sbtype=sb16vibra
irq=5
dma=1
hdma=5 [ne2000]
# If you want networking in Windows, set ne2000=true.
# This also requires that you set realnic= to a suitable value for your PC - that should be 1 for Raspberry Pi 4 wlan
ne2000=true
nicirq=10
realnic=1 [fdc, primary]
int13fakev86io=true [ide, primary]
int13fakeio=true
int13fakev86io=true [ide, secondary]
int13fakeio=true
int13fakev86io=true
cd-rom insertion delay=4000 [render]
scaler=none [autoexec]
imgmount 2 win98hd.img -size 512,63,130,1023 -fs none
# imgmount a win98boot.img -t floppy
imgmount d win98.iso Win98Soft.iso boot win98hd.img 

Boot your Windows disc

Enter the following at the DOS command prompt:

imgmount 2 win98hd.img -size 512,63,64,1023 -fs none
imgmount a win98boot.img -t floppy
imgmount d win98.iso -t iso -fs none -ide 2m
boot A:

Start Windows 98 Setup from CD-ROM, press ENTER to continue with setup as prompted, opt to configure unallocated disk space and enable large disk support, then press ENTER to restart.

DOSBox-X will restart its guest system and you – or your config file – should rerun the above mount and boot commands again.

Choose to start setup from CD-ROM again and the installer will format your blank disk image. Press ENTER when prompted to carry out the pre-installation check.

Install-time configuration

Agree to the EULA, then enter your Windows 98 Product Key. Click Next to accept the default C:\WINDOWS install directory. A ‘Typical’ install from the next set of options works for most users; select ‘Custom’ if you’re using a non-Latin character set.

Enter your name when prompted and move on to component selection, where you’ll find those additional character set options. Name your system, enter your network name, then set your keyboard, language, and location.

Skip creating a startup disk by clicking Next, then cancel when you’re prompted to insert a disc. The installer is finally ready to copy Windows 98’s files. Enable Turbo Mode from the DOBox-X CPU menu to speed this time-consuming process along. The guest system will reboot when finished.

Windows driver initialisation

At the DOBox-X Z: prompt, type:

boot win98hd.img

This boots into Windows 98 and, along with mount commands for the hard disk and a couple of ISO images, is already in our supplied win98.conf file, which you can use by launching DOSBox-X with:

dosbox-x -conf win98.conf

Windows 98 will take several minutes to initialise its drivers and may reboot during this process.

Set your time zone and network settings – the virtual network card’s IRQ should be set to 10 – when prompted. Setup will continue and the system will reboot.

At its first real boot, Windows will ask you to set a network password that will also be used for your system and then install more drivers from the CD.

Copy your install media

You should have functional sound, Voodoo graphics, and networking. Tell Windows that you’re using a LAN connection. If anything fails to install correctly, use the Add New Hardware wizard and Device Manager to find and configure the relevant items.

To make future driver installation easier, copy the /wing98 folder from the install CD image to your virtual hard disk using Windows File Manager. You can direct Windows to default to checking this directory in place of the CD by opening regedit, navigating to

HKEY_LOCAL_MACHINE\Software\Microsoft\Windows\CurrentVersion\Setup and changing SourcePath= to the location where you copied the files. That’s SourcePath=C:\\WIN98\ in our case.

Update your drivers

DOSBox-X allows you to mount multiple disc images using its imgmount command, and switch between them by pressing F12+LEFT-CTRL+C. Our config file includes an example and we’ve created an ISO image full of drivers, demos, and free software to get you started at The MagPi GitHub.

Install, in this order: 7-zip, Windows Installer 2.0 (instmsi), Internet Explorer 6, dotnet 1, dotnet 2, the visual C++ resdistributable, the K-Lite Codec Pack, Windows Media Player (MPClassic), and the Sound Blaster 16 driver.

Enable Turbo Mode to speed up the slowest installers, but switch to Normal speed before you type anything.

Install some software

Now you have a Windows 98 PC, you’ll want something to run on it. Our software disc includes some freeware and shareware applications, including Dink Smallwood and the first episode of Doom.

Online auctions can supplement your collection and some publishers have made parts of their back catalogue available for free. That includes Dynamix’s Tribes 2 mech combat sim, three of EA’s Command & Conquer RTS series and some of 3000AD’s Battlecruiser space sims.

Innoextract can extract Windows 98 compatible files from GOG packages, KernelEx runs Windows XP software, and Brasero can burn downloaded software to disc images for easy mounting. Install Brasero with:

sudo apt install brasero

Boot to Windows

If you want Raspberry Pi to really feel like a Windows 98 PC, you’ll want it to boot directly to DOSBox-X’s guest operating system. Open your config file, set fullscreen=true, and add any disc images you plan on using to the autoexec section. Also make a note of the shortcuts you’ll need to swap CD and floppy images.

mkdir /home/pi/.config/autostart
mousepad /home/pi/.config/autostart/dosbox.desktop

Add the following to the new text file and save it:

[Desktop Entry]
Type=Application
Name=DOSBox
Exec=/usr/bin/dosbox-x

Copy your config file to DOSBox-X’s default config file location:

cp /home/pi/win98/conf /home/pi/.config/dosbox-x/dosbox-x-0.83.3.conf

Now reboot Raspberry Pi into Windows.

Reading Time: 6 minutes

CrowPi2 is is a huge step up in ambition. The touchscreen is replaced with a full IPS display, keyboard, and a trackpad with mouse combo, turning the CrowPi learning environment into a laptop.

At the time of writing, CrowPi2 is in the last days of a successfully funded Kickstarter campaign, and the device will head to the Elecrow store shortly. The price reflects the Kickstarter and shipping cost.

There are three versions available: a Basic kit that contains the CrowPi2, a Starter Kit (which we are reviewing here) which also comes with a Raspberry Pi 4 with 4GB RAM, and an Advanced kit which comes with a 128GB microSD card and a few extra sensors.

The CrowPi2 itself is a plastic laptop clamshell with an 11.6-inch 1920×1080 IPS screen and a wireless keyboard (which can be removed from the housing and used separately). The case also comes with a 2MP camera, speakers, and a headphone jack. And a lot, lot more.

The CrowPi2 Development board

The real magic for CrowPi2 begins when you lift out the removable keyboard and take a look at the Development board beneath.

On the Development board you’ll find [“deep breath” – Ed]:

• Thumb joystick

• Segment display

• Relay

• LED

• DHT11

• Tilt switch

• LCD

• PIR motion sensor

• Sound sensor

• IR receiver

• I/O interface

• Servo interface

• Step motor interface

• 8×8 button matrix

• Buzzer

• Vibration motor

• Touch sensor

• 8×8 RGB LED matrix

• Light sensor

• Ultrasonic sensor

Alongside all that is a 40-pin GPIO socket and breadboard with a GPIO toggle switch that you can use to connect/disconnect all the components on the Development board and use pure GPIO I/O as regular.

There’s also a Raspberry Pi cooling fan (permanently on).

To be clear. This isn’t the list of components in the box. That’s just the permanently fixed electronics kit on the CrowPi2 Development board.

All the extra kit in the CrowPi2 box

The box contains a further 31 modules [“deep breath” – Ed], including:

• IR remote controller and receiver

• Stepper motor

• Servo

• SDC motor with mini fan

• RFID card and tag

• NFC card

• Moisture sensor

• Components pack (we see jumper wires, resistors, buttons, LEDs, a potentiometer, diodes, and capacitor.)

• Two game controllers

• Two 32 microSD cards: one with the custom Raspberry Pi OS system and another set up for RetroPie so you can quickly play games with the two controllers.

• Minecraft paper cut-outs to assemble (to make boxes and other elements from the game)

• User manual. Elecrow promises a Python and Scratch programming book to follow.

Honestly, it’s baffling the amount of electronics kit that’s on display here.

Learn code and electronics with CrowPi2

All this would mean little if it was simply components slung in a box. Thankfully, CrowPi has spent a lot of time honing version one, and CrowPi2 is a very well-rounded learning platform.

Our unit came with the Raspberry Pi 4 and a customised version of Raspberry Pi OS pre-installed. Setup was a case of plug in and turn on. We disassembled Raspberry Pi 4 from the case and fitted it back in to test the process, and it was very easy.

A custom version of Raspberry Pi OS boots into a friendly interface with links to projects, learning materials and various other friendly welcome features like Minecraft and AI examples. (You can boot into regular Raspberry Pi OS or other operating systems and still use the device).

Clicking the icon in the top right bounces you out to the regular Raspberry Pi OS desktop environment (and you can quickly click a Menu icon to get back to the CrowPi2 environment).

The tutorials are excellent. There are 16 Scratch tutorials and 32 Python lessons. Both make excellent use of the electronics components, having you build a flashing LED, automatic fan, tilt light – and, in the case of Python, working up to using sensors, RFID tags, voice recognition, and speech. A further AI section has five Speech Recognition courses and seven Face Recognition lessons (using OpenCV and the built-in camera).

It was as we worked through these courses that the CrowPi2 came alive in our hands and we decided it was something special, worthy of our recommendation.

Design decisions

Objectively speaking, CrowPi2 as a laptop is far from perfect and we’d be remiss not to mention the following (especially given our recommendation).
Some design decisions are questionable. The keyboard is flimsy with a lot of flex, the key response isn’t great and we shudder at the thought of typing out whole documents.

We’ve been worn down by countless terrible trackpads on Raspberry Pi laptops. While the CrowPi2 trackpad doesn’t disappoint us any more than pi-top 2 or NexDock 2, it’s still clunky. At least Elecrow included a mouse.

There are oddities like the camera USB cable that sticks out from the side to fit into one of the USB connections. There is no internal battery, and a storage compartment at the back doubles up as a power bank container should you want to work on the go.

On the side is a brightness/volume adjustment rocker. Confusingly, you press up first to adjust the brightness; down first to adjust the volume. The On button turns the device on, but not off.

Two things you’ll need to get started aren’t included: an AA battery for the mouse, and (less forgivable) a Micro-USB cable to charge up the wireless keyboard. Given the sheer weight of kit included, this is an odd choice.

The fan whirs constantly; The unit is almost two inches thick.

What we have here, however, is enough value for money and quality learning resources to make us overlook all of those flaws. Your mileage may vary though, and please buy with open eyes.

The laptop that isn’t one

Get past the idea of CrowPi2 as a laptop and it starts to make sense (see the NexDock 2 if that’s what you’re after).

After a while, we stopped thinking of it as a laptop and began to admire CrowPi2 as a compact electronics learning kit powered by Raspberry Pi 4. An electronics starter kit for the computer age.

It’s an all-in-one unit you can plug into a wall, put a kid (or willing-to-learn adult) in front of it, and let them get on with discovering how electronics components and computer code work together.

The learning materials go from simple Scratch up to advanced AI solutions. There’s a real journey inside CrowPi2 and we’re learning new techniques ourselves from it.

It’s the integration that impresses us: the combination of the Development board, all the components, and interactive tutorials. All in one place, all ready to go as soon as you connect the power.

Put it all together and you get a lovely piece of kit that we consider perfect for teaching and learning electronics and coding. We would have loved this to bits as kids. And let’s face it, that’s what Raspberry Pi is all about.

9/10

Reading Time: 2 minutes

It comes with an IP40 rating: “Protected from tools and small wires greater than 1 millimetre. Not protected from liquids.”

The modular cases come with either a solid lid or a side panel for an embedded display. We got three sizes to experiment with: 125×87 mm, 145×125 mm, and 237×195 mm

Made to measure

Matching accessories (which we haven’t tested) include stands, wall mounts, stacking adapters, and fibre-optic light guides – much of what you need to deploy multiple Raspberry Pi computers in a range of environments.

Inside the cardboard box, we received two half-shells, four side panels (with openings for the Raspberry Pi connectors), adhesive pads to stick the Raspberry Pi to the shell, and Torx screws.

We used a Torx T7 screwdriver to fix the adhesive pads to our Raspberry Pi 4 and a Torx T10 to screw the two outer shells together. The presence of Torx screws will help prevent prying hands from getting into the shell, and the microSD card is tucked safely inside.

Documentation is excellent, and a data sheet covers technical data and assembly instructions. On the whole, we like the build process and the case has an aesthetically pleasing style that lifts it above its industrial roots.

Verdict

8/10

Build quality is excellent and the versatility of applications will ensure the UCS Universal Housing finds itself happy in a wide range of industrial settings.

Reading Time: 3 minutes

What is your background in making and programming?

I graduated from a Japanese technical high school with a major in electric engineering. I have learned not only soldering to make electronic DIY, but also welding and programming with Apple IIe and punch cards.

I also came across the Sharp pocket computer PC-1416G (called Pokecon) as one of the learning materials at that time. It is for studying BASIC, though. I was trying to analyse its hardware to make some games with its assembly language. I investigated the internal system call to develop a mini graphical game with ‘poke, peek, and call’ commands, like IchigoJam.
I tore it down, looking into the hardware itself. It was really fun for me to play with it – I am sure that is why I love programming and making.

How did you learn about Raspberry Pi?

Frankly speaking, I was not so interested Raspberry Pi at first because I was still interested in OpenSolaris as a contributor. However, I saw it at an open-source event and tried it, and got interested and thought I would love to try to make some Linux servers with it. Now there are five original Raspberry Pi computers [used for] DLNA, MPD, Shairport (AirPlay), and Icecast streaming. They work really well. I’ve replaced one of them with Raspberry Pi 3 because of the lack of CPU power, though. Four of the Raspberry Pi boards are almost eight years old and work great.

What are the maker and Raspberry Pi communities like in Japan?

The Japanese Raspberry Pi community is still growing, but getting commoditised into business-to-consumer (B2C), [but] consumers often change their mind and B2C product life is really short lately.

On the other hand, the Japanese Raspberry Pi B2B business community is now growing rapidly. There are tons of use cases lately: AIoT, prototyping at an automotive company for self-driving, sensors at a vineyard, and more. Now I am planning to found a Raspberry Pi business community to help their businesses. There are also many more makers lately.

Do you have any involvement in other East Asian communities?

Yes, now I am concerned deeply with East Asian communities, especially helping the Taiwanese. CutiePi is the one of the projects I’m helping spread the word about. I will also help a couple of Taiwanese Raspberry Pi projects this year.

I have been attending COSCUP every year to expand and encourage their community. I am now staying in touch with some Taiwanese IT companies to use Raspberry Pi for their B2B business, and thinking of founding a Raspberry Pi business community there with them.

Supporting a community staying at home

“I am collaborating with the community in many ways: sending some official swag, holding events with them. Lately, almost all of the events are online or cancelled, and I have to find a way to help them […] – there are many more problems in communicating with others, especially as we cannot meet face-to-face. I hope the situation will get better and we can start growing again our Raspberry Pi community.”

Reading Time: 2 minutes

Dirk aimed to keep the build costs to around €300 – nearly two years later he estimates the project has cost him nearer €1500, though this includes the cost of two 3D printers and two previous prototypes. Now it’s built, there are no significant running costs. “Solar power is free and the built-in hardware is robust and reliable and has no real wearing parts, except the battery,” Dirk says.

False starts

The core parts are a Raspberry Pi Zero, Raspberry Pi Camera Module, and Witty Pi mini (a clock and power management module) plus GPS, solar panel, battery, motor, and motor controller.

As well as the Raspberry Pi OS and the Witty Pi software, Dirk made use of standard lines of code from earlier versions of his robot mower. He coded most parts of PiMowBot in REBOL3 since he’s been using the open-source functional programming language for nearly two decades. Python and GPIO Zero are used to communicate with the I2C devices, sensors, and motor control drivers.

Getting a precise orientation reading was a challenge due to the influence of magnetic north, but Dirk overcame this by getting PiMowBot to automatically go through a calibration process each time it’s set up.

He wanted to keep things as straightforward as possible, experimenting with various motors and drivers such as a Waveshare Motor Driver HAT to control the gearing, braking, and acceleration. However, the geared motors drew too much current, prompting a switch to an L298N dual H-bridge module. Further refinements included adding a more powerful motor, which also meant Tim needed to redesign the 3D case.

The final version of PiMowBot was complete by the start of this year, with build instructions and the software all available from Dirk’s website.

Mow-tivational

“The PiMowBot has two operating modes: an autonomous operating mode without induction wires in the ground, and a remote control mode to directly control the lawn-mower robot via its web UI and the location and orientation sensors on the lawn,” explains Dirk. He’s used the Camera Module for previous Raspberry Pi projects and is keen to further refine the optical object/obstacle detection with real-world testing during the summer months.

Although PiMowBot was intended as a “sophisticated hobby project” to work on in his and Tim’s spare time, Dirk is planning new models, including an XXL PiMowBot model and an intelligent system in which multiple PiMowBots operate alongside each other and share the mowing task. All hail our future mowing gods!

Reading Time: 5 minutes

Some people take things further and use a third-party service to help protect their Raspberry Pi. In this tutorial, we’re taking a look at remote.it to set up a work network using Raspberry Pi as a remote access gateway.

With remote.it, you can access a Raspberry Pi remotely without port forwarding. You do this by accessing the remote.it server as a proxy, or by using remote.it’s software to form a peer-to-peer network. Remote.it claims this is a safer way to set up a gateway than a traditional VPN.

Create an account

To connect to your Raspberry Pi remotely you’ll need to set up an account. Open the web browser and head to https://remote.it. Click on ‘Sign up’ and enter your email address. You will be emailed a verification code. Enter this code and pick (and confirm) a password. Make sure you choose a safe password (read the How To Geek guide on strong passwords).

Turn on SSH and VNC

To connect remotely, SSH and VNC must be activated on Raspberry Pi.

Click on the main menu and choose Preferences > Raspberry Pi Configuration. Choose the Interfaces tab and set both SSH and VNC to Enabled.

The VNC icon appears in the menu bar. Make sure you are not using the default password. (Choose the System tab in Raspberry Pi Configuration and click Change Password.) See the Raspberry Pi documentation ‘Securing your Raspberry Pi’ article for more information on keeping your computer safe.

Update and install

Make sure your Raspberry Pi is connected to the internet (using wireless LAN or an Ethernet cable). Open a Terminal window and enter the following commands:

sudo apt update
sudo apt install remoteit

When the packages are installed, the Terminal outputs the configuration information:

Continue your device configuration at http://find.remote.it or http://raspberrypi.local:29999 or http://192.168.0.41:29999

The IP (internet protocol) address will be unique to your network.

Register the Raspberry Pi

Open the web browser on your Raspberry Pi and navigate to http://find.remote.it. The window will search for devices running remote.it software (including your Raspberry Pi).

When your Raspberry Pi is found, the screen will display ‘Unregistered device’ along with the gateway details and IP address. (If this approach does not work, try one of the other two URLs listed in the configuration information above). Click on Register to add your Raspberry Pi to remote.it. You will be asked to enter the Email and Password for your remote.it account.

The Name field displays ‘raspberrypi’ (you can change the name to something else if you prefer). The Services window displays the three default services: ssh, vnc, and remoteit Admin Panel. Click Register again to add Raspberry Pi to your remote.it account.

Start a connection

You now connect to Raspberry Pi via remote.it. There are two approaches: via a web service, or using remote.it’s dedicated software.

First, let’s look at the web service. This provides a link to a proxy connection, and all data passes through one of remote.it’s proxy servers.

Open a web browser on your other computer and navigate to app.remote.it (enter your remote.it login credentials).

Click Connect and choose ‘Raspberry Pi – VNC’. The ‘VNC – Connection’ window appears, displaying host and port information.

Connect via VNC Viewer

Open your VNC software (we’re using VNC Viewer) and choose File > New connection. Cut and paste the combined proxy and port information into the VNC server field. Ours looks like this: proxy50.rt3.io:31249.

Double-click the icon and click Continue at the alert window. Enter ‘pi’ as the username and your Raspberry Pi OS password. Click OK to access the Raspberry Pi OS desktop remotely.

Connect via app

The drawback to the web service is that all your data passes through a proxy computer. For a more private connection, the remote.it app for Windows and macOS enables you to form a peer-to-peer (P2P) connection.

Data does not pass through remote.it servers (only the connection information), the data throughput is faster so the screen response time is better, and the connection will stay active as long as both devices remain connected. Finally, the URL remains the same and can be bookmarked (unlike the proxy connection which needs to be re-entered into VNC Viewer whenever you start the connection).

To create the P2P connection, you need to download remote.it’s desktop software. Click on ‘Get Desktop app’ from the app.remote.it website (or visit the download page) and click Download for Windows (or macOS, depending on your computer). Double-click the install file and follow the installation instructions. Open the remote.it software. You may see ‘We need to install or update our service in order to maintain background connections.’ Click on Install Service and Yes to the User Account Control alert.

Enter your remote.it account credentials into the app to sign in.

Under Devices, you will see your Raspberry Pi. Click it and click Connect next to VNC. The VNC icon will turn blue once connected.

You can launch VNC Viewer directly from the app (using the launch icon) or copy and paste the connection information into VNC Viewer (as in the previous step).

Cloaking your port

You can access Raspberry Pi remotely using remote.it. However, your VNC port is now open and listening. You can check this with the lsof (list of files) command:

sudo lsof -i -P -n | grep LISTEN

As well as the remote.it connection, it will show vncserver running on port 5900. The solution to this is to use a script, provided by remote.it, that cloaks this port. With the cloaking script active, people cannot scan your Raspberry Pi to find the open port because the port will not respond to incoming traffic. Meanwhile, remote.it can get through because remote.it software establishes the connection.

Upload scripts

You can use script remote.it with any language that runs on your Raspberry Pi. It’s best to practise with a test script. Head to the Uploading a Device Script page on remote.it’s website and click show-device-info.sh to download the file.

In the app.remote.it web interface, choose Scripting and click Upload. Change the option to ‘Executable script or program’ and click ‘Choose file’. Use the File Explorer window to find the file and click Upload.

Running the script

Scripts are run from the Devices window. Place a tick in the checkbox next to your Raspberry Pi and put a check next to ‘Show advanced columns’. (This will display the StatusA, StatusB, StatusC, and extra columns used by the test script). Click Actions > Execute Script; highlight show-device-info.sh and click Next.

Highlight the ‘Check to allow all devices to update the status columns during Job execution’ box. (This option displays information in the columns as the script runs.)

The cloaking script

Now download the cloak-vnc.sh script from remote.it’s GitHub page. Upload it in the same manner as the show-device-info.sh script. And run it from the Devices window.

This time you’ll get two options in the Bulk Execution window: ‘Cloak Service’ and ‘Uncloak Service’. Highlight ‘Cloak Service’ and click Finish and Submit. Scripts can take some while to run. Choose Scripting in the sidebar and you will see the script progress. You can also cancel and delete scripts from this window.

On average it takes three minutes for the job status to update. So your script may actually complete and update the cells before the Job Status catches up. It should respond within three to four minutes.

Cloaked and connected

Now connect to your Raspberry Pi via the remote.it app for a secure connection. Your port will be cloaked on Raspberry Pi and because you’re connecting via remote.it’s service, you will not be sharing your IP and port information publicly. And the P2P service ensures that your data is not flowing through remote.it’s service.

Reading Time: 3 minutes

Many robots use the two wheels of a DC motor to provide tank-like steering. PiCar-V 2.0 takes things up a gear with an additional servo motor to steer the front wheels via a rack mechanism (called the Servo Linkage Plate).

A further two servos provide a pan-and-tilt mechanism for the included USB webcam (not an official Raspberry Pi Camera Module, unfortunately).

You have to bring your own Raspberry Pi board and microSD card to the table, plus two 18650 rechargeable Li-ion batteries (3.7 V) and a charger. Even so, there’s a lot here for your money.

Days to build

Oh, my days, there are so many items in the box. By our count, there are 128 separate components, including four wheels, three servos, two DC motors, the chassis, battery back, three HAT boards, and a wild amount of screws, standoffs, and nuts.

The very first step for each front wheel involves the wheel, a flange bearing, three connector plates, three more hex fixing plates, an M4×25 screw, and an M4 self-locking nut.

Everything you need is in the box, including a screwdriver, socket wrench, and two small wrenches. Fortunately, there’s an 88-page full‑colour booklet to walk you through the build and usage.

Even so. This is a taxing build. The assembly guide is available online as a PDF, so why not take a good look through it to gauge your comfort factor with the build process before buying?

We found the build straightforward, if fiddly. The two exceptions being attaching the front wheels to the steering mechanism, and screwing on the plates to the rotating servo mechanism (without rotating the servos themselves).

It’s mostly time-consuming. Like jigsaws, Lego, and Meccano kits, there’s a lot of fun to be had in long and detailed builds.

The setup regarding Raspberry Pi OS is a little outdated (recommending NOOBS or Balena Etcher, while Raspberry Pi now recommends the purpose-built Raspberry Pi Imager tool. But the steps still work, and it walks you through setting up Raspberry Pi OS, enabling SSH, and connecting to the PiCar-V robot over your local network.

Bring it together

The result is a good-looking car, if imperfect. The design prevents access to the HDMI port once PiCar-V V2.0 is assembled, so you can’t attach a screen to debug Raspberry Pi OS if you lose the SSH connection. The design also means that the physical build is interrupted halfway as you set up the software on Raspberry Pi.

Once complete, the frame is bendy and, in truth, we’d prefer the Upper Plate to have a little more heft. We didn’t encounter any problems in use, however.

There’s also a nest of wires between the boards, and the USB camera cable extends over the front-left wheel. Some cable ties will come in handy.

Software support is extensive and everything is available on the GitHub repo. You’ll need to install Python and use pip to install PyQt5 and the requests library on Raspberry Pi.

When all is set up, you can run the server on Raspberry Pi and run the client software on your PC. The client software displays the webcam view, and you control the rover with W, A, S, D and the camera pan and tilt with the arrow keys. Android and iOS apps are available.

There is an example Python program that uses OpenCV to track a red ball. SunFounder has also provided a tutorial for programming the robot with Dragit (a Snap-based visual programming tool). You can read this, and lots of other documentation on the SunFounder Learn site. When everything is finished and you’re done with the example code, there are eight digital and four analogue channels on the Robot HAT, plus two spare I2C ports. Plenty of space for expansion with extra sensors.

Verdict

8/10

We like this kit, but it has its flaws. You get a lot of components and some pieces of kit should be of a higher quality. It requires patience, but the supporting documentation is excellent. A nice robot car with potential for expansion.

Reading Time: 2 minutes

Raspberry Pi 4 web hosting with Mythic Beasts

Hire a Raspberry Pi in the cloud and run web services at twice the speed and half the cost! Mythic Beast’s new Raspberry Pi 4 web hosting service is now live.  

Run Windows 9x software

Boot a Raspberry Pi into Windows 98 and run classic software.This tutorial from KG Orphanides is a great way to rediscover computing from the 1990s.

Make making accessible

Ensure that everyone can join in with the fun of your builds by making them accessible to the widest possible audience.

Masafumi Ohta interview

Masafumi founded ‘Japanese Raspberry Pi Users Group’ in October 2012, It’s thanks to the efforts of people like Masafumi that we’ve been able to see some of the amazing work from makers in Japan and even other parts of East Asia.

We deliver to your door

Buy The MagPi magazine issue #96 from the Raspberry Pi Press store and we will deliver it straight to your door. Plus! Take a 12-month subscription in print and we’ll give you a free Raspberry Pi Zero computer and starter kit worth £20.

Reading Time: < 1 minute

Save 37% off the cover price with a subscription to The MagPi magazine. Try three issues for just £5, then pay £25 every six issues. You’ll save money and get a regular supply of in-depth reviews, features, guides and other PC enthusiast goodness delivered directly to your door every month.

Subscribe

Reading Time: 3 minutes

The Girls Into Coding (GIC) events were incredibly successful, so Helene has decided to commit fully to the organisation: “Earlier this year, we set higher ambitions and established GIC as a Community Interest Company aiming to offer more girls more opportunities to develop.”

When did you first learn about Raspberry Pi and CoderDojo?

I first learnt about the Raspberry Pi about five years ago at a Raspberry Jam in Covent Garden, Central London. Someone at a Code Club meetup recommended going, so Avye and I did. Avye really enjoyed going to these kinds of events so we continued to go to other Jams. They encouraged her to craft, code and design, and allowed me to discover new things.

At one in particular, I was approached by the organiser of a CoderDojo and was invited to become a mentor. I accepted and soon volunteered as a Scratch mentor. Since my introduction to Raspberry Pi, I’ve enjoyed tinkering with the various iterations, and last year I attended the Picademy in Cambridge to become a Raspberry Pi Certified Educator.

What is Girls Into Coding?

Girls Into Coding was an idea that Avye had when she was ten. By this age she had moved from attending workshops at CoderDojo to preparing and delivering them herself at the Kingston University Dojo and at other community events. After repeatedly observing that most attendees at her workshops were boys, Avye identified a need to help more girls to access these events and was determined to encourage more girls to give STEM a go. As a result, she decided to set up ‘Girls Into Coding’(GIC), a series of events targeting girls.

Girls and women are under-represented in STEM and the new generation has a chance to change that. Girls Into Coding is our way of contributing to that change. Moreover, GIC was set up to encourage girls to explore and enjoy STEM subjects and to encourage them to pursue further activities, education, and careers.

We provide learning opportunities for girls aged 10 and 14. We’ve created workshops and education events to immerse young people in a range of STEM activities. Our workshops are led by Avye and specialist STEM educators, and assisted by volunteers. The workshops consist of robotics, physical computing, and 3D printing. As well as participating in hands-on workshops, the events give the girls an opportunity to listen to lightning talks throughout the day, delivered by inspiring female role models who are doing cool stuff in the tech world.

Do you have any advice for anyone wanting to mentor or help teach computing subjects?

I would encourage anyone to be a mentor, even if they do not have the computing knowledge at first. You can learn as you go along. Preparation is key. Spend some time preparing the session in advance, give yourself enough time to get used to the content, and practise.

Mentoring is a very rewarding experience. It can be daunting at first but after a few mentoring sessions, your confidence will grow and before you know it you will soon find out that you can do it and have fun in the process! Be brave and do not overcomplicate it! There are lots of online tutorials that are available for those starting out. They are a very good place to start.

Reading Time: 3 minutes

“Anyone can come by and see the stills run and learn about them in more detail,” says son Jim, cheerfully. Given its location on a rather pretty island just north of Seattle, Washington, and the quality of the liqueurs and whiskeys produced there, it’s little wonder the Heisings have plenty of visitors. Their blackberry liqueur, distilled from one of Bev’s recipes, is the highest-rated of all spirits in North America by Tastings.com. The other special ingredient in the business’s success is Raspberry Pi.

Distilling whiskey with Raspberry Pi

Although he’s responsible for the monitoring equipment and tech that keeps it running, Jim says it’s really a ‘mom and pop’ business. His mum comes up with all the recipes while “dad figures out how to adapt our stills and manufacturing process to produce them in large quantities.”

Steve is a former aerospace engineer, while Jim has more than 20 years’ experience as a self-taught software engineer and startup founder. He built most of the hardware and software for the still, which dad Steve designed. “Hardware is relatively new for me, but I learned enough to get by with tinkering and asking a lot of questions,” reveals Jim.

Distillation doesn’t require much technology to run, Jim explains, but it’s very labour-intensive and using technology can make a big difference to what a distillery can do. “Many smaller distilleries struggle because they spend so much time distilling that they have little time for anything else,” he says. Often, they will modify alcohol bought in from a larger concern and put their own label on it. The Heisings wanted to make their own.

Raspberry Pi enables their stills to run completely automated 95% of the time. Jim explains that its CPU is powerful enough and has plenty of RAM (he’d love direct 12 V DC power via the header pin too, though). Other setups he’s used couldn’t. “We have 15+ temperature sensors, a barometer, multiple D2As (digital-to-analogue converters) connected to high-voltage variable SSRs (solid-state relays), multiple low-voltage D2As for pumps, and multiple relays for solenoids.”

“Having I/O systems we could communicate with through USB was a bit of a game-changer,” he says, “because we could run and debug the whole control system from a laptop, then simply unplug the USB cable and plug it into Raspberry Pi to let it run completely hands-off.” Jim credits “amazing” I/O products from Tinkerforge. “We connect Raspberry Pi via USB to Tinkerforge’s products, import a simple SDK, then we have an enormous set of options for I/O at our fingertips.” You can see the source code and wiring diagrams at GitHub.

Ever evolving still

The Whidbey Island setup is still evolving, says Jim. “Every day we’re making changes and we continue to build new stills for our distillery. At some point in the future, we may even offer kits for other distillers to make their own, but only if we can find some time (which is always hard).”

Reading Time: 2 minutes

Recently, VueScan released a version of its software that works on Raspberry Pi, so we decided to check it out on Raspberry Pi OS.

Problem solver

Getting VueScan is pretty easy – head to hamrick.com and it will recommend you a version to download for the OS you’re using. There are instructions on how to install the version from this download; however, we recommend hitting the link for other versions and finding the ARM32 DEB file for installation. It’s a bit easier, only involves a single file, and will then appear in the program menu under Accessories. VueScan can be used for free, although it will embed a watermark onto your scans. It’s a good way to test if your scanner will work with it, though, especially as we had some issues with a scanner that was listed as being supported.

Troubleshooting was quite easy, however. Common problems are listed and, once you’re through that shortlist, you’re taken to the website to find out more. The developers behind the software are a contact form away from giving you some support on the software.

Once you have the scanner working, while you can just hit Scan and get a good picture, you have a lot of options to play with as well. Any function the scanner has, you can change, and you can even add some post-processing to the scanned images: fixing colours, making the imager sharper, and even allowing for character recognition (OCR) so the text of the image can be highlighted. Your mileage might vary on how well that OCR works, though – it’s highly dependent on how good your scanner is and the item you’re scanning.

Verdict

8/10

It’s a great piece of software that makes scanning on Raspberry Pi much easier than it’s ever been.

Reading Time: 6 minutes

Forked from the original DOSBox emulator, DOSBox-X has more precise hardware emulation, supports a wider range of software, and can effectively run more DOS-related operating systems (up to Windows ME). It also has a sophisticated graphical interface to help you manage tasks such as configuration and virtual disk-swapping.

DOS emulation: you’ll need

Create your DOS directories

Let’s create the directory structure that we’ll use to house the software we’ll run through DOSBox-X:

mkdir -p dos/{floppy,cd,games}

The floppy and cd directories will house disk images which we’ll be able to switch between in DOSBox-X. This tutorial and our template config files presume you’ll keep everything in a /home/pi/dos/ directory, so be sure to change any paths if you’re using a different username or dos directory names.

While our generic config file should handle most DOS software well on Raspberry Pi, you can also create separate .conf files for specific programs to better match their requirements and automatically run commands.

Tweak your graphics

Assuming you’re using a standard 1920×1080 display with your Raspberry Pi, you’ll find some more demanding DOS software struggling at full resolution, particularly if you have DOSBox-X configured to use OpenGL and aspect ratio correction.

On the desktop, open the main menu, go to Preferences and select Screen Configuration. Right-click on your display – most likely marked HDMI-1 – and from the Resolution menu, select 1280×720. Running your entire GUI at a lower resolution will lighten the load of rendering and upscaling for the emulation and have no adverse effect on games from an era when 800×600 was the norm.

Install DOSBox-X

In a Terminal, enter the following:

sudo apt install automake libncurses-dev nasm libsdl-net1.2-dev libpcap-dev libfluidsynth-dev ffmpeg libavdevice58 libavformat-* libswscale-* libavcodec-*
git clone https://github.com/joncampbell123/dosbox-x.git
cd dosbox-x
./build
sudo make install
dosbox-x

DOSBox-X should open at its Z: prompt. You can’t paste commands into it from the clipboard, but there are some modern convenience features: tab auto-completes, you can scroll through your command history using the up arrow, and you can add startup commands to a config file.

Type exit to quit and ensure that the config directory, which we’ll need in the next step, is created properly.

Export a config file

Restart DOSBox-X and tell it to generate a config file that we can later modify in a text editor, based on the program’s default settings and then exit.

CONFIG.COM -all -wcd
exit

The file we’ve just made can be found in /home/pi/.config/dosbox-x and, at time of writing, is named dosbox-x-0.83.3.conf.

As well as being human-readable and conveniently editable in a text editor, you can modify this long and extensively commented file from within DOSBox-X using the configuration GUI in the main menu. This is handy, since DOSBox‑X’s configuration has more options than that of vanilla DOSBox.

Customise your config

For this tutorial, we’ve created some config files that you can download from The MagPi GitHub page. The code box will run most DOS software. As well as editing your main DOSBox-X config, you can launch DOSBox-X with a specific config file – useful if you wish to easily switch between different OS setups – using the following command-line switch:

dosbox-x -conf yourfile.conf

We’ll take advantage of that later to help install Windows 3.11. Note that your custom config files need only include lines that vary from the defaults. In the following steps, we’ll create a config file optimised for playing late-era DOS games on Raspberry Pi 4 with 4GB or 8GB RAM.
Graphics, scalers, and performance

The default config is already well optimised to run DOS software on most systems, but we should make a few adjustments to improve performance on Raspberry Pi’s hardware.
Leave the fullscreen setting as false, as you can enable and disable fullscreen mode using DOSBox‑X’s menus or the F12+F keyboard shortcut; fullresolution should be left as ‘desktop’.
To get proper aspect ratio correction and reasonable graphical fidelity at 1280×720, you should set the output to opengl, aspect to true, and select a scaler up to interpolate low-res graphics. Scaler choice is largely a matter of personal taste, so use the Video menu options to try a few. If your sound becomes choppy, you’re pushing Raspberry Pi’s capabilities too far.

autoexec.bat

At the end of the config file is autoexec, where we’ll put all our MOUNT and IMGMOUNT lines to assign drive letters to directories and floppy or CD images, and any commands to run at boot.

In our sample config, we’ve used MOUNT to set /home/pi/dos as DOS’s drive C. It’s here that we’ll copy and install all our software to.

If you use the IMGMOUNT command with multiple file names of CD or floppy images, you’ll be able to swap between those images in order to swap between media. To swap floppies, use F12+LEFT-CTRL+D. To swap CD or DVDs, use F12+LEFT-CTRL+C.

Using DOSBox-X

Like DOSBox, DOSBox-X uses the open-source FreeDOS operating system, rather than Microsoft’s proprietary MS-DOS. Navigation through directories isn’t too different to using a Bash terminal, particularly as a number of Bash commands have been included, such as LS as an alternative to DOS’s DIR, and CD .. alongside the usual DOS CD.. command. To run a .exe, .com, or .bat file, just type its name without the extension.

To capture and release your mouse, use the LEFT-CTRL+F10 shortcut. The autolock entry under SDL config enables capture-on-clock.

Buy Windows 3.11

Now we’ll install Windows for Workgroups 3.11, released in December 1993. The biggest challenge is finding a copy to install – usually ageing floppy disks, or disk images if you made backups. We’re working from a set of disk images.

If you don’t already have one and don’t fancy the second-hand market, you can, surprisingly, find it included in Microsoft Visual Studio Subscriptions (formerly MSDN Subscriptions), currently priced at £33.54 per month, for the benefit of developers working on backwards compatibility.

Install Windows

Copy the contents of each installation disk or image to a /win311 subdirectory of the dos directory tree we made earlier; you can do this as you normally would on the desktop or at the command line, or by using DOSBox-X’s IMGMOUNT to mount them and using the DOS COPY command while switching disks. At the command line, start DOSBox-X with a Windows-suitable config file – download ours from GitHub.

dosbox-x -conf win311.conf
CD WIN311
SETUP

Windows 3.11 will install itself. Reboot.

CD WINDOWS
WIN

Using Windows 3.x

If you’ve only ever used Windows 95 or later, Windows 3.x’s interface may feel somewhat alien. There’s no Start button and if you want to quit back to the DOS prompt, you have to open Program Manager’s File menu and select Exit Windows…

The default Program Manager folders, each full of shortcuts to helpful software and settings, are clearly labelled. To explore your mounted DOS drives, open Main and then File Manager. Accessories include MS Paint precursor Paintbrush, a Sound Recorder, and even a Media Player.

A line at the top left of each opened window allows you to move and close it, while minimise and maximise buttons are at the right.

Boot Raspberry Pi to DOS

Once you’ve configured DOSBox-X – and any relevant window managers – to your satisfaction, you can complete your pitch-perfect 1990s PC simulation by booting straight to DOS. Open a Terminal window and type:

mkdir /home/pi/.config/autostart
mousepad /home/pi/.config/autostart/dosbox.desktop

Add the following to the new text file:

[Desktop Entry]
Type=Application
Name=DOSBox
Exec=/usr/bin/dosbox-x

This will use DOSBox-X’s default config file. You’ll need to enable fullscreen in your DOSBox-X config for this to launch correctly, and opengl-dependent aspect ratio correction is also strongly advised.

Reading Time: 3 minutes

It took him around four months, on and off, to develop his Piano-Playing Robot. As shown in his YouTube project video (), it features 15 solenoids mounted on a wooden frame placed on top of an electric piano keyboard.

“Basically, the system is quite simple,” reveals Étienne. “Each solenoid pushes on a piano key and is controlled by a specific GPIO pin of a Raspberry Pi 3 Model B.” Since the latter doesn’t have enough electrical power to activate the solenoids directly, an electromechanical relay module acts as a buffer between the two.”

Fast ‘fingers’

The system can read a MIDI file, convert it into on and off signals for each note, and play it. A basic UI allows the user to choose between scales, arpeggios, or melodies.

“As of now, the robot can play a full scale in just under two seconds,” reveals Étienne. “That is, at the fastest rate, it plays about eight keys per second.” This impressive speed makes it possible to get the timing of the notes just right.

“However, the system is limited by the push and pull times of the solenoids, which add up a few tens of milliseconds,” he says. “At a first glance, it may not seem like much, but in a fast melody it can make a difference. Our robot still has a long way to go to overtake human-beings: the fastest pianist in the world claims to be able to reach up to almost 20 keys per second!”

Sight reading

So far, he has programmed the robot to play six scales, six arpeggios, and five different melodies. The most impressive feature, however, is its ability to read previously unseen sheet music – using a Raspberry Pi Camera Module – and play it.

“A camera was added to the system to read and decode new musical scores that had never been shown to the robot before,” says Étienne. “It then scans, decodes, and plays them.”
To do this, it uses a free open-source tool called Audiveris. “It is a complex program for optical music recognition (OMR), which is recognition of musical symbols. Unlike optical character recognition (OCR), this task requires decoding the configuration of symbols in two dimensions. In addition, the order of magnitude of the symbols is extremely wide (sometimes ranging from tiny dots above musical notes to symbols that are drawn across a large part of the page).”

In brief, the recognition process is done with classifiers and neural networks that make matches with symbols previously learned. “For my project, the decoding part was done on a laptop since Audiveris was more easily compatible with Windows,” he explains. “On average, it takes around 30 seconds for the robot to decode a sheet of music containing a few staves.”
Étienne is hoping to add extra features to the robot. “In particular, I would like to extend the concept to 88 solenoids. This way, the robot would theoretically be able to play all the existing melodies on the piano! Obviously, there is a lot of development to do in terms of Optical Music Recognition, but I like to imagine a piano concert performed entirely by a robot!”

Reading Time: 5 minutes

Attach Inky wHAT to Raspberry Pi

Carefully line up your Inky wHAT over the GPIO header pins on Raspberry Pi (use the header booster that sits between the wHAT and the board if yours is a full-sized Raspberry Pi model) and press it into place. Connect Raspberry Pi’s mouse, keyboard, screen, and power on. As with using any new hardware, start by updating your Raspberry Pi. We advise using a fresh installation of Raspberry OS, which will suggest updating itself when it first boots up. Follow the prompts to check for and download any updates.

Turn on SPI and I2C

Inky wHAT needs both SPI (Serial Peripheral Interface) and I2C (Inter-Integrated Circuit) enabled to work. Both can be found in Raspberry Pi Configuration. Click on the menu icon and choose Preferences > Raspberry Pi Configuration. Click on the Interfaces tab and set both SPI and I2C to Enabled. Click OK.

Install Inky wHAT software

With the Inky wHAT hardware attached, we’ll install the software. Open a Terminal (click the black rectangle icon in the top bar) and enter:

curl https://get.pimoroni.com/inky | bash

Press Y in response to ‘Do you wish to do a full install?’, then press ENTER.

Clean and reset

During the wHAT installation, some example programs will have been copied to Raspberry OS. Open a new Terminal window and type:

cd /home/pi/Pimoroni/inky/examples
ls

…to locate them. Let’s start by ‘cleaning’ your Inky wHAT’s screen (it ships with the logo displayed.) While in the examples directory, enter:

python clean.py --type what --colour yellow

(If you have the red wHAT, then enter –colour red instead, and for all following examples.)
The Inky wHAT screen flashes as it refreshes and then cycles through yellow, black, and white. In the end you will have a clear screen. To bring back the fantastic Inky wHAT logo screen, type :

python logo.py --type what --colour yellow

Quotes

Inside the examples directory is another folder marked what. There are more examples in here, including a selection of quotes. Enter:

cd what

…to change directory. Then run the example:

quotes-what.py –colour yellow

For these programs, you don’t need to specify the type (because you’re in the what folder). The quotes are pulled from an editable selection at Wikiquote.

Change quote

You can edit the quotes-what.py code to pick quotes from any of your favourite people. Enter:

mousepad quotes-what.py

…to open the file in Text Editor. And replace the list of names with your favoured person:

people = [ "Douglas Adams"
]

Use File > Save As and save the file as quotes‑adams.py. Now, in the Terminal, enter:

python quotes-adams.py --type yellow

…and you’ll see a quote from the celebrated author.

Make a name badge

Now we’ve explored the options Pimoroni provides, let’s tell Inky wHAT to display something we specify. Head back to the examples directory, using cd .. or cd /home/pi/Pimoroni/examples, and type:

python name-badge.py --type what --colour yellow --name "The MagPi"

Replace ‘The MagPi’ with your own name (or any name you like). After a few seconds, Inky wHAT’s display will mimic a conference delegate’s name badge, complete with the name you provided.

Hello Python

What about coding your own text and images from scratch? Fortunately, it’s possible to code your own messages using Python. Let’s start with the traditional Hello World program. Open the Thonny IDE (Menu > Programming > Thonny Python IDE) and save the file as hello_inky.py (we saved ours in the Pimoroni/inky directory).

Now enter the code from hello_inky.py. Click Run to display Hello World on the screen.

Show off your own images

You can tell Inky wHAT to display any image you wish, but it can only handle 400×300-pixel PNG files that use three-channel indexed colour.

Although specific, it doesn’t mean you can’t display most images, just that you will need to convert them first. The GIMP image editor is ideal for this. To install GIMP, type:

sudo apt update
sudo apt install gimp

You will also need the three-channel indexed palette. Either use Text Editor to create the inky‑palette.gpl file and save it with your code files (we’ve put ours in /home/pi/Pimoroni/inky), or download it from Pimoroni’s Inky GitHub repo – in a Terminal, enter:

git clone https://github.com/pimoroni/inky

You’ll find the colour palette at inky/tools/
inky–palette.gpl. It’s configured for the Inky wHAT red version, so change the third line from 255, 0, 0 to 255, 255, 0 if you have a yellow Inky wHAT.

Import the palette

Once installed, type gimp in the Terminal to run GIMP, or open it from Menu > Graphics > GNU Image Manipulation program. Choose Windows > Dockable Dialogs > Palettes. The Palettes dialog appears in the top-right of the screen. Right-click it and choose Import Palette. Choose Palette File and click the Select Palette File icon next to it. Use the window to select the inky-palette.gpl file. Click Import.

Convert an image

You can use inky-palette.gpl to convert an RGB image to the three channels. For best results you can create your own image using GIMP (Choose File > New and set the width to 400 and the height to 300). And then draw an image using the tools and stick to the three colours that match your Inky wHAT model (white, black, and yellow/red). If your artistic skills aren’t feeling up to much, we’ve included a MagPi-Logo.png file along with the code files on our GitHub. It’s already sized at 400×300 and features the three colours.

Open the image (File > Open) and click Convert to the colour profile (this isn’t related to our three channels). Now choose Image > Mode > Indexed and select Use Custom Palette. Click the palette graphic (typically it will be marked Web) and choose inky-palette from the list.

Click Convert.

Now choose File > Export As and save the file to the same directory as your code (we saved it to /home/pi/Pimoroni/inky).

Image code

The code to display an image on the Inky wHAT is much shorter, and easier to understand, than the code for text.

Enter the code from magpi_inky.py and make sure that the file name and path on line 7 (magpi_logo.png) match the file you saved in the previous step. Run the program to view the image on your Inky wHAT:

python magpi_inky.py

Reading Time: 3 minutes

“This March, two friends of mine organised a Raspberry Jam in my hometown, Amsterdam.” Jolien explains. “There I saw some really cool projects. Living in Amsterdam, more living space in my apartment is always very welcome and I’ve wanted a loft bed for some time.”

Instead of simply modifying an existing bed, Jolien went the extra step and built her own, using just a jig-saw and drill as that’s all she had.

“I built a double loft bed in a tree-house style with LED lights and an embedded display with a clock and the weather forecast,” she says. “The control panel also has three buttons which can be used to turn the lights on and off. The light strands I bought were connected to battery packs, so I cut those off and wired the strings to a Raspberry Pi.” The bed is approximately 3 m by 1.7 m and its height is 2.6 m, while its steps also function as shelves for the ground floor.

Loft Bed: Quick Facts

• Jolien tied the safety nets herself

• It all runs off a single Raspberry Pi (for now)

• There’s approximately six metres of light strips

• The control panel buttons are arcade buttons

• Weather forecasting was a later addition

Building and wiring

“When I started building, I wasn’t sure what features I wanted to add and Raspberry Pi is very flexible,” Jolien recalls. “It’s also very easy to set up SSH, which was necessary since the electronics are embedded in the side panel and can’t be reached directly.”

For just a bed, it’s got some fairly advanced tech in it, as Jolien reveals: “Besides the control panel in the bed, I also made a progressive web app which enables me to turn the lights on/off and set an alarm via my smartphone. If the alarm is triggered, Raspberry Pi turns on the lights and a song is played. Everything runs on Python and I used Python Flask to make an API on Raspberry Pi so that the GPIO pins connected to the lights can be turned HIGH and LOW from the outside.”

Sleep on it

It sounds amazing already, and we’d probably put down some serious cash for a product version of it. However, Jolien isn’t quite finished yet. “I think this project took me about 100 hours in total,” she says. “At this moment, the speaker is not attached yet; I will do this next week. I also want to design a 3D-printed laptop wall mount for the ground floor.”

We’re eager to see what else Jolien can come up with to add to an already very impressive project.

Reading Time: 3 minutes

“BB is a small, wall-powered robot designed to test out offline computer vision, detecting people and faces,” says Sean.

“It searches a room for people, then tries to recognise any faces if it detects a person. If it comes across a person in its memory, it will respond appropriately, e.g. ‘Hello Sean’, in a squeaky voice. It tracks any people it detects, following them around the room with its pan-tilt mounted camera.”

BB Robot: Quick Facts

• BB has an 180-degree viewing angle

• Like PiArm and other arm robots, BB cannot move around

• There’s an inner case that holds all the electronics

• Sean designed a relay circuit based on this one

• The high-pitched voice is supposed to replicate the turrets from Portal

Seeing robot

Like a lot of Raspberry Pi robots, BB uses OpenCV. It’s a Python library that we’ve covered in the magazine before that allows people to add machine learning to image processing. Computer vision works fine on a Raspberry Pi – OpenCV has been used for the Formula Pi racing league, and face unlock Raspberry Pi tutorials have been around for as long as Raspberry Pi has.

See: Raspberry Pi face recognition

Sean tells us he has a lot of experience using a Raspberry Pi:

“I’ve found it a great platform for building robots with. It is more powerful than Arduino and other microcontrollers, yet smaller and cheaper than full desktops.”

A Raspberry Pi 3A+ is the main component, but extra circuitry is included. “A DIY power management circuit is used to turn Raspberry Pi on and off with a button, with a power indicator LED,” he explains. “An amplifier is included to power a small speaker, as well as connectors for the two servos. The camera is a standard Raspberry Pi [Camera Module], and the whole robot is held together with a black 3D-printed case, with a white shell to cover unsightly bolts.”

First steps

Creating BB was just the start of what Sean plans to do, with bigger and better robots already in development.

“I wanted to make a small 3D-printed robot, since I just acquired a 3D printer and wanted to make something cool,” he says. “I also wanted a platform to test offline computer vision on. My long-term goal is to create a Raspberry Pi-powered robot, capable of computer vision, speech recognition, and speech synthesis. I’ve already finished building my new platform to recognise speech… It is cube-shaped, with a similar colour scheme to BB. It features four microphones, as well as a speaker and LED ring. I’m hoping to combine the two robots to create a sophisticated machine, as if from a movie.”

We look forward to seeing this sophisticated machine in the future.

Reading Time: 2 minutes

You gotta roll with it

The system is built around a Raspberry Pi, a HiFiBerry AMP to output high-quality sound to a hidden speaker, and an Arduino Mega to take readings from eight Hall effect sensors. In each toilet roll holder, a 3D-printed part contains a circular four-pole magnet that triggers the Hall effect sensor when rotated. Whenever it receives data from the sensors via the Arduino, Raspberry Pi outputs the respective sound sample for each moving toilet roll.

“As long as a roll is rolling, the Arduino will send a continuous message telling the Pure Data patch to play that specific sample,” says Max. “It’s built in such a way so that the voice will sing as long as the roll is rolling.”

A few technical issues were encountered in the build process, including random instances of failure to work: “My audio patch (program) consumed 100% CPU and clogged the whole Raspberry Pi,” recalls Max. “That version used windowing functions to create seamless pitch shifting of the samples I used for audio playback. Had to abandon that…”

Max also had to fiddle with the Hall effect sensors for quite a while before he was happy with their placement, and a colleague assisted with the 3D prints, but all their efforts proved worthwhile.

When it comes to changing the rolls, this is just done normally and the only thing that people need to remember is to put the roll back in the correct direction: “The magnets inside the 3D print must face the Hall effect sensor on the left.”

Flushed with success

While Max has had great feedback on his creation, “there hasn’t been much interaction since people are out of office nowadays. I’m currently learning how to play it so I can hold toilet concerts.”

Integral to the Singing Toilet, and to all his work, has been Raspberry Pi. “[It’s] my go-to tool for interactive audio, mainly because I’m so used to it, but also because there’s such a great community where any help needed is out there to be found,” Max tells us. “My past projects with Raspberry Pi include a huge dance carpet, a synth made of steel, museum exhibits, and various interactive audio projects.”

Reading Time: 2 minutes

Supplied fully assembled, it’ll work with any 40‑pin Raspberry Pi model. Pimoroni has thoughtfully included six metal PCB standoffs (with screws) for spacing and securing it, and/or the attached mini or full-size HATs, via pairs of the eight mounting holes provided.

Pin checking

No extra software is required. So, in theory, you just mount your two mini HATs on the HAT Hacker HAT and you’re all set to go. Well, you do need to check that the same pins aren’t used by both HATs (unless they’re I2C pins with different addresses), which is easily done by looking them up on pinout.xyz – or using the pHAT Stack Configurator, although the latter offers a limited selection of HAT options and could do with updating.

If you need to attach more than two HATs to Raspberry Pi, a pHAT Stack would be a better option, as you can attach up to five mini HATs or three full-size HATs.

For most project setups, however, the HAT Hacker HAT should suffice. You might well want to connect just one mini HAT, e.g. a display, and use the other set of pins for connecting electronic components – for which the full labelling of the second header’s pins is very useful. Unlike the similar FullpHAT, however, there are no bonus GPIO breakout holes in addition to the two main headers.

Verdict

8/10

A compact and solidly built solution to mounting more than one mini / full-size HAT on Raspberry Pi at a reasonable price