thisismyrobot

My first 3D printable model!

2018-03-10T16:01:00.002+11:00

I recently purchased an RX5808 Pro Diversity FPV receiver and wanted to mount it on my FPV box-style goggles in a manner that didn't just involve a metric ton of double-sided tape.

I've been interested in 3D printing for a long time, but the cost for a "turnkey" 3D printer has remained just out of reach. On top of this I hadn't done any 3D modelling since my Year 12 CAD class over fifteen years ago.

Thankfully, that first problem can be resolved by places like Shapeways and the second by a truly awesome program called OpenSCAD - a programming-driven 3D modeller.

A day's furious measuring and coding, followed by two weeks delivery time from the US of A and look what arrived:

Even better, it works:

I've tweaked the control slot a touch since these photos, so if you just want the STL file, it's in my drones repository, or you can support me by purchasing it from my Shapeways shop.

Installing OpenCV 3.3.1 with Python 3.5 bindings on a Raspberry Pi Zero running Raspbian Stretch

2017-12-30T11:54:00.000+11:00

I've been experimenting with some basic computer vision concepts on a Raspberry Pi as part of a small mobile robot project. Fundamental to this was getting OpenCV compiled and set up in Python, a process that took 5 or 6 tries over the majority of my Christmas/New Year break. Hopefully this guide will save you some proportion of that time when you have a go at it...

Assumptions
I'm working off the assumption that you are using Python 3.5 and are working within an activated virtual environment for your project. I am also going to use "~/myproject" as a stand in for your specific project path. This means you will have already done something like:

You need to stay in that activated virtual environment for the duration of this guide.

Caveats
This install will be missing ffmpeg due to a bug in the current version.

Virtual memory
I needed to add more virtual memory to get this to work, using the following steps:

Installation
Now, the fun part. These are the steps you need to run, in order. The 'make' step will take 6-ish hours so is a good candidate for leaving overnight.

Virtual memory - again
It is important that you undo the virtual memory change above or you will risk corrupting you SD card. Just revert the value to 100 and start and stop the service.

From here you should be able to import 'cv2' from within the virtual environment's Python, and get on with actual development :)

This guide couldn't have been made without the efforts of:

NERFrame build write-up

2017-07-09T12:50:00.003+10:00

My NERFrame FPV drone has generated enough interest that I've decided to do a build write-up - so without further ado here it is!

Drones drones drones

2017-07-02T16:06:00.000+10:00

I've got a new hobby! A couple of months ago I purchased my first FPV drone, the entry level Eachine QX95.

This glorious little machine came with a Betaflight-compatible flight controller, and I've since learnt how to build the ARM-targeted firmware (a first for me) after adding a software failsafe.

I'd played with an FPV camera on a basic toy drone before but this is my first foray into a FPV-specific drone. So far I've been learning fairly quickly

And then I pranged it.

Which became the perfect opportunity to build my first drone "frame", out of old NERF rounds and hot glue :)

The aptly-named Pink Rocket flies suspiciously well, though it's a touch docile compared to the original due to all the wind resistance from the cross-sectional area. It'll certainly do until the replacement frame arrives.

Like most of my hobbies, there's a GitHub repo.

Temporarily switching Wi-Fi networks from the command line on a Raspberry Pi

2017-01-28T10:51:00.001+11:00

I recently had a scenario with my Raspberry Pi (Zero, running Raspbian) where I needed to temporarily switch away from the primary Wi-Fi connection to another access point (to drive a camera) then reconnect to my original network. Turns out wpa_cli can do the trick:

The best thing about this approach is that it is non-persistent, so if it all goes horrible wrong you get your original networking configuration back upon reboot.

Open sourcing all the things

2016-06-18T11:06:00.002+10:00

A while ago I was blogging about GenghisIO - a project with the goal of removing all the mess involved in programming a small range of real-world robots - the IDE, drivers, cables etc etc.

The project eventually stalled due to increasing complexity and decreasing resources (time...) but quite a few neat problems were solved along the way - so I have open sourced it!

If you're interested in using Google App Engine applications, Python Flask with WebSockets, empythoned, Python QR code generation, Android WebViews and JavaScript -> Bluetooth communication then go and have a look.

I'd love to hear from you below if you find something useful in there :)

Wireless headless Raspberry Pi Zero with APC220 radios

2016-04-30T15:10:00.000+10:00

As a bit of a weekend project and as a way to (finally) use the really cool APC220 Radio Communication modules that I bought ages ago I've put together a wireless and "headless" setup for my Raspberry Pi Zero. This allows me to have a remote-controllable Pi Zero placed anywhere within 1 km (line of sight) without wires! I'm hoping to use this as a robot backpack initially, probably for the Jumping Sumo.

I won't go into technical detail as the build was simply connecting the power and TX/RX from the Pi UART to the APC220 module (I've used protoboard to make a "hat") and telling the Pi to use ttyAMA0 for the console in /boot/cmdline.txt.

If you have any specific how-to questions, let me know in the comments. In the meantime enjoy the photos :)

The top of the hat

The bottom of the hat and Pi header

At the local end I am using an USB-TTL adapter and Putty to connect to the Pi.

It's all in the (domain) name

2016-02-06T10:39:00.002+11:00

After taking a crash course in CNAME entries and CloudFlare's free SSL I've moved dnstwister to a shiny new domain:

dnstwister.report

Using CloudFlare has restored 99% of the performance-helping caching that the old Google App Engine site had - the only negative from the migration.

The DNS entries are still propagating at the time of posting so if you have any issues you can still use the old domain.

DNSTwister is now on Heroku

2016-02-03T18:47:00.000+11:00

Well, the title says it all. I've moved dnstwister to Heroku.

It didn't take a lot of work and the benefits are huge - the killer for me was the ability to control exactly how much the application scaled up cost-wise. The costing for hobbyist use on Google App Engine was, in my opinion, completely unmanageable. You couldn't limit the performance of your app under load, the only option was to just pay more and more until you covered the quota exhaustion. With Heroku if I get a surge in interest it'll just get a bit slower, and I can choose to scale up at that point as necessary.

I also like deploying through pushing a branch to GitHub and the whole admin dashboard interface is really slick.

Bookmarked report links will automatically redirect from the old domain so you shouldn't feel a thing and the performance is right on par with Google App Engine.

Enjoy!

Driving a DFRobot 3-Wire LED module from a Raspberry Pi

2015-12-26T19:03:00.000+11:00

Here's some very simple test code that you can use to drive one of these LED modules from a Raspberry Pi:

(The misspelling was intentional, I figured "Raspbery" looked better than "Raspberr" :D)

Pic: Just need to fill in the missing bits #boxingday #shopping #littlebirdelectronics

2015-12-26T10:29:00.001+11:00

Pic: #walle

2015-12-18T21:08:00.001+11:00

DNS Twister updated

2015-12-16T19:57:00.000+11:00

DNS Twister has been updated to fix a number of small bugs, add a few features (it's significantly faster) and - for the sake of your eyes - there's been the addition of a bit of css.

Go have a play!

Side project time: DNS Twister

2015-11-08T18:05:00.000+11:00

Recently I came across a really awesome Python tool called dnstwist. In the author's words, dnstwist is a "Domain name permutation engine for detecting typo squatting, phishing and corporate espionage".

At its core dnstwist creates permutations of domain names and checks to see if they are registered. It also does a ton of other cool things like GeoIP mapping to resolved IP addresses.

As I am sure there are people who would like to use dnstwist but don't have Python installed, I have (after a weekend's re-familiarisation with Google App Engine) created DNS Twister!

It may look rough as **** but it works, presenting the core dnstwist functionality via a web application.

The source code is on GitHub of course.

Docking/charging system overview

2015-10-25T10:10:00.000+11:00

As a follow-up to yesterday's post, here's another of my ~~excellent~~ terrible Visio diagrams, showing how the different bits of software will work together on the RPi:

Click to see in all it's glory

The SumoProxy component is effectively complete and the Web Interface component allows viewing the FPV camera but doesn't offer any control. SumoPy and the docking code "work", but not yet in the manner shown.

Sumo-charge update

2015-10-24T21:18:00.001+11:00

Thanks to a super-long-weekend my Parrot Jumping Sumo wireless charging project has been making good progress.

The current design has the Raspberry Pi "proxy" the Jumping Sumo, teeing off the data passing between the controller (iPad/phone/etc) and the Sumo.

There is a basic web interface where the user can view the FPV video, and (soon) enable the automated docking sequence. The web interface runs off a flask web server which can parse the video data and will inject controls back into the stream.

I expect I'll switch to web-only control once I've proved the web interface can perform on the Raspberry Pi.

SumoPy and SumoProxy
Two components of this project are available separately via these repositories if you want a play: SumoPy, SumoProxy.

Faster barcode tracking

2015-10-18T16:32:00.000+11:00

Leveraging the work done to speed up both video capture and motor control, here's a significantly faster tracking example:

A video posted by Robert Wallhead (@thisreallyismyrobot) on Oct 17, 2015 at 9:18pm PDT

On board it looks like this:

Success!

2015-10-18T13:44:00.001+11:00

The first photo taken by properly scraping the video stream.

Capturing Parrot Jumping Sumo video packets in Python's UDPServer

2015-10-18T10:09:00.000+11:00

One of the key goals of my sumo-charge project has been to have the Sumo auto-dock by driving towards a "landmark" (a barcode in my case).

To do this I obviously need access to the video stream from the Sumo. The video UDP packets are the same format as the rest of the response data, but for some reason I was not receiving them using Python's UDPServer.

After much flailing around on Google I fired up Wireshark and discovered those packets were being sent still, so it was definitely a problem with my UDP server. Digging deeper (thanks, stack overflow) I discovered that the maximum packet size in UDPServer was hard-coded to less than the size of the video packets.

The size of the video packets sent by the Sumo is established in the reply during the initialisation handshake to TCP port 44444, in the 'arstream_fragment_size' field (65000 in this instance):

To receive the video packets you need to update your UDPServer instance's 'max_packet_size' with the same value:

Pic: It's a while since I last used Visio...

2015-10-16T19:08:00.001+11:00

Pic: Interested parties #cats

2015-09-26T13:37:00.001+10:00

Pic: Low tech motor calibration #sumo

2015-09-23T20:04:00.001+10:00

QR tracking using the Parrot Jumping Sumo camera

2015-09-21T17:21:00.000+10:00

Nested QR code with the Sumo's wheels against the monitor.

To implement "auto-docking" I am intending to use a 7x7cm square nested QR code target and the Sumo's FPV camera. The QR nesting allows detection and recognition at distances up to 50cm whilst remaining accurate right up the moment of touching the docking station. The pictures shown here are actual snapshots from my Jumping Sumo's FPV camera (640x480 resolution).

Same QR code image at an approx range of 30cm from the monitor.

I've prototyped some detection code using an install of the python-qrtools package on the Raspberry Pi:

This code successfully recognises the two QR codes in these test images, though it takes about half a second to do so.

Next up, a docking control loop!

Pic: Drone life #alwayscharging #sumo

2015-09-20T09:24:00.001+10:00

Capturing photos from a Parrot Jumping Sumo using Python

2015-09-19T11:01:00.000+10:00

Success!

After a slightly faulty first effort I have successfully taken a photo from my Jumping Sumo using Python's telnetlib module.

In the following code the built-in yavta tool is used to push (and buffer) a snapshot from the camera to /dev/stdout and then base64 is used to push an encoded version of /dev/stdout back to the Python client where it is decoded and saved to disk. As before I have to kill the dragon-prog process to gain access to the camera.

There are a number of alternative image capture routes using yavta, e.g. saving the file to the Sumo's flash or a ram disk and using FTP to recover it. This one seems to be the quickest (and simplest) for now and will suffice for the purposes of identifying and homing in on the charging station.