Merge pull request #147 from orionrobots/import-2018-build-journals

First piwars 2018 build journal converted - video linked too.

Author: dannystaple

content/2017/05/07-skittlebot-build-journal/2017-05-07-skittlebot-build-journal.md

@@ -0,0 +1,148 @@
+---
+title: Skittlebot build journal, 7th May 2017
+date: 2017-05-07
+tags: [robot building, piwars, opencv, electronics, power, raspberry pi robotics project, raspberry pi opencv project, gpiozero robotics project]
+thumbnail: content/2017/05/07-skittlebot-build-journal/skittlebot-chassis-thumb.png
+---
+Skittlebot is designed using the 4Tronix initio robot base. It is built around a Raspberry Pi for remote control, and for experiments with computer vision.
+
+I'd built this initially as a simple Arduino robot in 2016, but now it is time to make it something far more exciting. It's called Skittlebot because it's original goal was to follow a plastic, brightly coloured skittle (bowling pin) around. 
+
+It is being prepared for possible entry into the PiWars 2018 competition.
+
+This setup used VNC to control the robot and see the output of the camera. The camera is a Raspberry Pi camera module.
+
+## Progress so far - OpenCV colour tracking
+
+Last night the opencv work to track colour was finally attempted right on the robot itself.
+
+I was able to get the output I wanted - console output, and the display (via vnc) of the tracking, with the tuning parameters.
+
+However, vnc is a little bit iffy here, and stops actually doing screen updates.
+
+I then started to combine it with code to drive the motors.
+The motors and the shared power were unplugged, with the Pi running off a USB  supply while I was working on it.
+
+I also seem to have destroyed a memory card yesterday on this set up.
+
+When it tried to drive, the system reset - after a tiny movement of motors.
+
+So - the next step is to look for power issues.
+
+The power issues could be:
+
+- Low batteries - check them all - should add to more than 6.6v.
+- Bad connections - check these - I did undo wiring.
+- The UBEC could actually have gone wrong, as I had USB voltage while the UBEC still present. lets hope not.
+- Under-volting as motor current makes the system drop - the UBEC should help, but it drops at 6.6v.
+
+## Batteries
+
+This is unloaded voltages on an array of 8 AA batteries - these are high voltages for NiMH.
+
+- 1.363
+- 1.364
+- 1.365
+- 1.365
+- 1.365
+- 1.365
+
+About 8v. They are dead or dying alkaline batteries.
+
+## Wiring
+
+Nothing there seems to be wrong, or shorting.
+Lets put batteries in again.
+
+Measure this, and try the simple motor test code.
+-8.18? probes are right. Battery box red terminal on red probe. Whaaat?
+
+Okay - just confusing wiring.
+Powers up - motors turn until I disable motor controller
+Lets see the Raspberry Pi come up...
+
+VNC has reconnected.
+
+Lets try initialising it (I should initialise to stopped early)
+Can I measure the battery output now? 7.34 - still good. (no it wasn't, they were dead)
+
+## Testing
+
+Using gpiozero robot object in the code.
+
+- Initialised.
+- Motor power enabled.
+- Lets try forwards.
+- Reset? Or non responsive after enabling power. Lets switch from vnc to putty.
+- It's not reset - uptime 5 minutes.
+- Just vnc..
+
+Okay it moves.
+So that is not the reset.
+
+- `robot.forward()` was fine.
+- `robot.left()` was a reset!
+
+Why is that?
+
+- Uptime confirms it is a reset
+- Backwards is fine.
+- Right is ALSO a reset. Is this a big current pull trying to turn? Can we reduce current by reducing speed? The l298 is famously not too clever about this overcurrent stuff.
+
+## Making a lower power board
+
+Lets rejig power:
+
+Ubec to Pi and for L298 digital side. Only motors go from battery directly.
+Create a lower power distribution board (regardless of battery/usb it makes sense).
+
+From bottom to top deck we need:
+
+1.  +5v
+2.  gnd
+3.  line - right - signal
+4.  line - mid - signal
+5.  line - left - signal
+6.  motor left a
+7.  motor left b
+8.  motor right a
+9.  motor right b
+10. motor sense left a
+11. motor sense left b
+12. motor sense right a
+13. motor sense right b
+
+call it 14 (extra one for a sensor?)
+
+Then we bring up 2 sets of 7.
+
+### Right side
+
+Pi Power is here
+
+1. +ve
+2. gnd
+3. motor right a
+4. motor right b
+5. odometry right a
+6. odometry right b
+7. line sense right
+
+### Left side
+
+1. unusued
+2. line sense middle
+3. motor left a
+4. motor left b
+5. odometry left a
+6. odometry left b
+7. line sense left
+
+Wiring is getting too messy.
+Lets swap it out for a picon zero - sounds better.
+
+## Next steps
+
+- Get the wiring working and reliable.
+- Get the power sorted.
+- Make a video of this in action

content/2017/05/07-skittlebot-build-journal/skittlebot-chassis-thumb.png

@@ -0,0 +1,40 @@
+---
+title: My Raspberry Pi computer vision robot Skittlebot works!
+date: 2017-05-14
+tags: [robot building, piwars, opencv, electronics, power, raspberry pi robotics project, raspberry pi opencv project, gpiozero robotics project]
+thumbnail: content/2017/05/14-skittlebot-video/skittlebot-working-screenshot.png
+---
+I now have my Raspberry Pi based  computer vision demonstration robot working! This is a big step forwards, as I can start looking at more interesting projects and competitions with it.
+
+## Specifications
+
+- 4Tronix PiCon Zero
+- Raspberry Pi 3
+- V1 Camera module
+- 4Tronix Initio chassis
+- Logic power is a USB power bank
+- Motor power is 6xAA batteries
+- Raspberry Pi image created with PiBakery - and using precompiled NUmpy and OpenCV libraries
+- Programmed in Python
+
+## Updates since last time
+
+I've rewired to use the PiCon Zero instead of the L298n and all that point to point wiring. This has made it a lot neater, and there's less to debug.
+
+I've also separated the Raspberry Pi power out to a USB power bank. This means I can run the Pi without the motors, and I can also run the motors without the Pi. This is a good thing, as it means I can debug the motors without the Pi crashing. It also means that large current issues won't reset the Raspberry Pi.
+
+## Video
+
+The Skittlebot video tells the story of it's creation, and adaptation from the Arduino robot, along with the first tests of the computer vision system.
+
+<iframe width="560" height="315" src="https://www.youtube.com/embed/z14HcflsRW0" frameborder="0" allowfullscreen></iframe>
+
+Skittlebot was built when I found that armbot was too big to come to PyCon Uk 2016. It uses a camera and visual processing to locate coloured objects, and knock them over. See how I built it, see it running, why I made it and how the code was written.
+
+The video mixes footage that was unreleased from multiple sessions, and includes the point where I'd shaved my head for a while!
+
+## Code
+
+The software uses PiCamera, Python and OpenCV.
+
+The code is at <https://github.com/orionrobots/skittlebot>.

content/2017/05/14-skittlebot-video/2017-05-14-skittlebot-working.md

@@ -0,0 +1,36 @@
+---
+title: Turning Skittlebot into a PiWars Robot
+date: 2017-10-13
+tags: [robot building, piwars, opencv, electronics, power, raspberry pi robotics project, raspberry pi opencv project, gpiozero robotics project, piwars, robotics competitions]
+---
+Skittlebot based on the Initio chassis may actually be a little slow to be a piwars bot, but it's certainly a great platform to get the basics down.
+
+So the plan is a set of experiments, to get the code going, get the Raspberry Pi and power in a good state, then look at making it all faster and lighter.
+
+this plan is likely optimistic, but lets get as many ideas down as we can first.
+
+## Major areas to cover
+
+* Remote - I have a Playstation Six Axis controller. Getting this paired with the device, and repeatable so I have code + PiBakery recipe is expected.
+* Straight Line autonomous - build a rig similar to the straight line at piwars - get SkittleBot good at touching no sides and going as fast as it can along the line - not that there IS a line to follow on the floor too - which may be easier than other sensors.
+* Camera - the over the rainbow challenge, and the rear ultrasonic sensor.
+* Spec out replacement chassis - customised:
+    * 3D printed, off the shelf, toy hack or CNC carved?
+    * Faster motors.
+    * Line follower modules.
+    * Power - separate Pi and motor (as SkittleBot already does).
+    * Make the bottom half only first - get this right, with a simple bracket for the pan/tilt mechanism I already have.
+* Test driving WS2801's from a Raspberry Pi.
+* Replacement top
+    * Get styling here - something unique to orionrobots - possibly doing a few thick structural 3D print areas, but many thin/hollow areas that are purely for style.
+    * Consider the WS2801 6 Status LED's at this point.
+
+## Remote
+
+The SixAxis controller is Bluetooth. There is a [PiBorg guide](https://www.piborg.org/rpi-ps3-help) on connecting them up. Lets follow that - and get it driving over bluetooth.
+
+Consider the inputs library? <https://pypi.python.org/pypi/inputs>
+
+A bit less heavy than pygame - but not sure it'll work with the ps3 sixaxis. This library has come from Zeth, a rather instrumental part of PyCon UK!
+
+The PyGame route: [PS3 controller with PYGAME](https://www.raspberrypi.org/forums/viewtopic.php?f=32&t=147234)