Saturday, July 14, 2018

Raspberry PI, PWM, servos, and PCA9685

The code mentioned below can be found in this git repo.
Pulse Width Modulation (PWM) is the technique used from a digital source to simulate an analog output.
For example, imagine that you want to dim an led from a digital device, to make it look like it is glowing. The digital device only has pins that can take 2 values: 0 or 3V3.
0 means that the led will be off, 3V3 means it will be on, at 100% of its brightness.
In short, it is on or off, and there is nothing in between.
But here is an idea to work around that issue:
To show it at 50% of its brightness, the idea is to turn it off 50% of the time, and on 50% of the time.
To show it at 25% of its brightness, it will be on 25% of the time, and off 75% of the time.
If the on-off cycles are short and fast enough, a human eye will no be able to see them, it will only have the illusion of the resulting brightness.
A human eye cannot make the distinction between images separated by less than one 10th of a second. That is why the movies are shot at 24 images per second, so you cannot tell the difference between the frames.
This technique is call Persistence of Vision (POV).
The #1 parameter of PoV is the human retina. To have an idea of how much it is important, just put your cat in front of a TV, and see how much he/she reacts. To a cat, it might just be a fuzzy screen...
The early movies - like Charlie Chaplin's silent ones - were shot at 16 images per second, fast enough to induce POV. They were later projected by faster projectors - 24 frames per second. That is why the characters seem to move faster. They were originally moving normally.


Here are examples of PWM applied to POV:
At work, for real:
The PCA9685 is a servo driver PCB.
The Raspberry PI does not have analog pins, we need to use Pulse Width Modulation to simulate analog values, a servo is an analog device.
We use for that the method setPWM(channel, 0, pulse), that will eventually write to the registers of the device.
An instruction like setPWM(channel, 0, pulse) means:
  • On channel channel (0 to 15 on the PCA9685)
  • in each cycle, turn the power on between 0 and pulse.
pulse has a value between 0 and 4095, that is 4096 distinct values, 4096 is 212, the PCA9685 is a 12 bit device.

The frequency

The frequency is provided in Hertz (Hz). A frequency of 60 means 60 cycles per second.
At 60 Hz, a cycle will be 1 / 60 second, which is 0.01666666 second, or 16.66666 milli-second (ms).

The pulse

For each of the cycles set above by setting the frequency, we need to determine the int value, between 0 and 4095, corresponding to the pulse in milliseconds we want to simulate with PWM.
In the class i2c.servo.pwm.PCA9685.java, this is done in this method:
public static int getServoValueFromPulse(int freq, float targetPulse) {
  double pulseLength = 1_000_000; // 1s = 1,000,000 us per pulse. "us" is to be read "micro (mu) sec".
  pulseLength /= freq;  // 40..1000 Hz
  pulseLength /= 4_096; // 12 bits of resolution. 4096 = 2^12
  int pulse = (int) Math.round((targetPulse * 1_000) / pulseLength); // in millisec
  if (verbose) {
    System.out.println(String.format("%.04f \u00b5s per bit, pulse: %d", pulseLength, pulse));
  }
  return pulse;
}
The cycle length (in ms) obviously depends on the frequency.
The pulse required for the servo to work is emitted once per cycle.

Example

As an example, let us calculate for a 60 Hz frequency the pulse value to send to setPWM(channel, 0, pulse) for a 1.5 millisecond PWM:
  • 1 cycle has a duration of 1 / 60 second, or 16.66666 milliseconds.
  • each cycle is divided in 4096 slots, we can say that 4096 bits = 16.6666 ms.
  • the solution is provided by a rule of three: value = 4096 * (pulse / 16.66666), which is 368.64, rounded to 369.

A comment about servos' compliance and reliability

Theoretically, servos follow those rules:
PulseStandardContinuous
1.5 ms0 °Stop
2.0 ms90 °FullSpeed forward
1.0 ms-90 °FullSpeed backward
That happens not to be always true, some servos (like https://www.adafruit.com/product/169 or https://www.adafruit.com/product/155) have values going between 0.5 ms and 2.5 ms.
Before using them, servos should be calibrated. You can use the class i2c.samples.IntercativeServo.java can be used for that, you set the pulse values interactively, and you see what the servo is doing.
$> ./inter.servo
Connected to bus. OK.
Connected to device. OK.
freq (40-1000)  ? > 60
Setting PWM frequency to 60 Hz
Estimated pre-scale: 100.72526
Final pre-scale: 101.0
Servo Channel (0-15) : 1
Entry method: T for Ticks (0..4095), P for Pulse (in ms) > p
Enter 'quit' to exit.
Pulse in ms > 1.5
setServoPulse(1, 1.5)
4.0690 μs per bit, pulse:369
-------------------
Pulse in ms > 0.5
setServoPulse(1, 0.5)
4.0690 μs per bit, pulse:122
-------------------
Pulse in ms > 0.6
setServoPulse(1, 0.6)
4.0690 μs per bit, pulse:147
-------------------
Pulse in ms > 2.4
setServoPulse(1, 2.4)
4.0690 μs per bit, pulse:589
-------------------
Pulse in ms > 2.5
setServoPulse(1, 2.5)
4.0690 μs per bit, pulse:614
-------------------
... etc.

Once you have determined the appropriate min and max values, you also have the int values to feed the setPWM with.

Some links:

Tuesday, June 12, 2018

Languages Comparison

For the fun: Same problem addressed in several languages, read the paper here.

It is about matrix and systems of equations resolution, curve smoothing, etc.

Done in C, Java, Processing, Scala, Kotlin, Python, JavaScript, Groovy, Go, Clojure (in progress), ...

Tuesday, April 24, 2018

Controlling invisible machines with emails, from Java

Here is the problem

You have your network at home, with several machines connected to it (laptops, tablets, Raspberry PIs, phones, etc). Your home network is a Local Area Network (aka LAN), the machines can see each other, but they cannot be seen from outside, from the Internet.
You may very well want to deal with those machines while away from home, to restart services, launch a new program, or even reboot.
In the configuration mentioned above, this is simple, you just cannot do it. And that is frustrating!
There is a way though. Those machines on your home LAN can send and receive emails...

Using JavaMail

JavaMail is a Java package that has been available for ever, it understands the email protocols (IMAP, POP3, SMTP, etc), and can be used to interact with email accounts programmatically.

An example

There is an example of such an interaction on this github repository.
The fastest way to get it running is to run the following commands (these are for Linux - and MacOS - on Windows, use the git shell):
$ git clone https://github.com/OlivierLD/raspberry-coffee.git
$ cd raspberry-coffee
$ cd common-utils
$ ../gradlew shadowJar
$ cp email.properties.sample email.properties
$ vi email.properties
$ # Here you modify your properties file to match your email account
$ java -cp ./build/libs/common-utils-1.0-all.jar email.examples.EmailWatcher -send:google -receive:google
The -send:google -receive:google depends on the settings in your email.properties.
Then, to the account mentioned in the email.properties, send a message like this:
Subject: execute
Content:
whoami
ifconfig
uname -a
Note: this example requires the content to be in plain/text.
Once the message is received by the EmailWatcher, it sends you an acknowledgement:
Then, the 3 commands are processed by the EmailWatcher, you would see in its console an output like that:
Start receiving.
Received:
whoami
ifconfig
uname -a

Operation: [execute], sent for processing...
pi
lo0: flags=8049 mtu 16384
 options=1203
 inet 127.0.0.1 netmask 0xff000000 
 inet6 ::1 prefixlen 128 
...
And finally, you receive an email like that:
... meaning that the commands you've sent have been executed.

You can also attach the script to execute to a blank email, with topic execute-script:
Attach a file like this:

#!/bin/bash
whoami
ifconfig
ps -ef | grep EmailWatcher
... and just wait for the result to come back to you:
Scripts execution returned: 
pi
eth0: flags=4099  mtu 1500
        ether a4:ba:db:c9:04:2e  txqueuelen 1000  (Ethernet)
        RX packets 0  bytes 0 (0.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 0  bytes 0 (0.0 B)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
        device interrupt 18  

lo: flags=73  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10
        loop  txqueuelen 1  (Local Loopback)
        RX packets 9215  bytes 2022884 (1.9 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 9215  bytes 2022884 (1.9 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

wlan0: flags=4163  mtu 1500
        inet 192.168.42.3  netmask 255.255.255.0  broadcast 192.168.42.255
        inet6 fe80::4038:1f53:b94f:ccc2  prefixlen 64  scopeid 0x20
        ether 78:e4:00:78:ad:8f  txqueuelen 1000  (Ethernet)
        RX packets 8848724  bytes 696021134 (663.7 MiB)
        RX errors 0  dropped 0  overruns 0  frame 18848059
        TX packets 6040965  bytes 795472510 (758.6 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
        device interrupt 17  base 0xc000  

pi      12476 12472  1 16:39 pts/0    00:00:53 java -cp ./build/libs/RasPISamples-1.0-all.jar weatherstation.email.EmailWatcher -send:google -receive:google
pi      16204 16199  0 18:04 pts/0    00:00:00 grep EmailWatcher
>> sh ./attachments/2018-04-26_18-04-27/sample.sh returned status 0

Comments

This process is not synchronous, this could be a drawback... But still, it allows you to interact remotely with machines invisible from the Internet.

Having the command

java -cp ./build/libs/common-utils-1.0-all.jar email.examples.EmailWatcher -send:google -receive:google
fired when the machine boots will allow you make sure it is waiting for your emails as soon as the machine is up.

This EmailWatcher as it is also allows you to execute scripts, attached to the email. Look into the code for details ;)
It is even possible to ssh to another machine and execute a bunch of commands stored in a script... The command you send in the email's body would be like

ssh pi@192.148.42.13 bash -s < ~/nodepi.banner.sh
If a password is required, use sshpass:
sshpass -p 'secret-password' ssh pi@192.148.42.13 bash -s < ~/nodepi.sudo.sh
You can even sudo:
echo 'secret-password' | sudo -S privilegedCommand
This can be dangerous, hey? With great power come great responsibilities...

Sunday, April 15, 2018

Head-Up Display (HUD)

The idea here is to display a screen on a transparent support - like a wind shield.
The data are displayed on the screen, reflected on the transparent support, and nothing is preventing you from seeing through it.
(Click the image to enlarge it)

Here is above an HTML page, tweaked by some CSS classes to mirror the data (as the page is reflected on the screen, the page content has to be displayed as in a mirror, and flipped upside down.). In this case, the page is rendered on Chromium in kiosk mode, running on a Raspberry PI with a touch screen attached to it.
CSS Classes:
    .mirror {
      display: block;
      -webkit-transform: matrix(-1, 0, 0, 1, 0, 0);
      -moz-transform: matrix(-1, 0, 0, 1, 0, 0);
      -o-transform: matrix(-1, 0, 0, 1, 0, 0);
      transform: matrix(-1, 0, 0, 1, 0, 0);
    }

    .upside-down {
      height: 100%;
      width: 100%;
      -moz-transform: rotate(180deg);
      -webkit-transform: rotate(180deg);
      -ms-transform: rotate(180deg);
      -o-transform: rotate(180deg);
      transform: rotate(180deg);
    }

    .mirror-upside-down {
      display: block;
      -webkit-transform: matrix(-1, 0, 0, 1, 0, 0) rotate(180deg);
      -moz-transform: matrix(-1, 0, 0, 1, 0, 0) rotate(180deg);
      -o-transform: matrix(-1, 0, 0, 1, 0, 0) rotate(180deg);
      transform: matrix(-1, 0, 0, 1, 0, 0) rotate(180deg);
    }

In the picture above, we use the class as follow:
<div id="the-div" class="mirror-upside-down big" style="padding: 0px; text-align: center;">
  <hr/>
  <table>
    <tr>
      <td colspan="2">GPS Data</td>
    </tr>
    <tr>
      <td>
        <span>Your position:</span>
        <br/>
        <span>N 37° 44.93'</span>
...
The page on the screen (not on the wind shield) would actually look like this:

GPS Data
Your position:
N 37° 44.93'
W 122°30.42'
Your Speed:
12.34 kts

You can also work around the perspective effect on the reflected page by tweaking the CSS classes:
    .mirror-upside-down {
      display: block;
      -webkit-transform: matrix(-1, 0, 0, 1, 0, 0) rotate(180deg) perspective(50em) rotateX(-40deg);
      -moz-transform: matrix(-1, 0, 0, 1, 0, 0) rotate(180deg) perspective(50em) rotateX(-40deg);
      -o-transform: matrix(-1, 0, 0, 1, 0, 0) rotate(180deg) perspective(50em) rotateX(-40deg);
      transform: matrix(-1, 0, 0, 1, 0, 0) rotate(180deg) perspective(50em) rotateX(-40deg);
    }

GPS Data
Your position:
N 37° 44.93'
W 122°30.42'
Your Speed:
12.34 kts
We call this the Star Wars effect. ;)

Possibilities are endless!
The full page is here.

Wednesday, April 11, 2018

Docker on the Raspberry PI

This post intends to illustrate how Docker can work around the "But it works on my machine!.." syndrome.

Let's say you have a nodejs project you want to share with others.
The application reads GPS data through a Serial port, and feeds a WebSocket server.
The data can then be visualized through a Web interface.

To enable everything, you need to:
  1. Have a Raspberry PI
  2. Flash its SD card and connect it to a network
  3. Install build tools
  4. Install git
  5. Install NodeJS and npm
  6. Clone the right git repository
  7. Install all the required node modules
  8. Drill down into the right directory
  9. Start the node server with the right script
  10. Access the Raspberry PI from another machine on the same network, and reach the right HTML page.

This is certainly not difficult, but there are many ways to do several mistakes at each step of the process!

Docker can take care of the steps 3 to 9. It will build the image, and then run it.
The image can also be pushed to a repository, so users would not have to build it.
Just to run it after downloading it.

The only pre-requisite would be to have installed Docker on the machine (the Raspberry PI here), as explained here.
Create a Dockerfile like this (available here):
 FROM resin/raspberrypi3-debian:latest

 LABEL maintainer="Olivier LeDiouris <olivier@lediouris.net>"

 RUN echo "alias ll='ls -lisah'" >> $HOME/.bashrc

 RUN apt-get update
 RUN apt-get install sysvbanner
 RUN apt-get install -y curl git build-essential
 RUN curl -sL https://deb.nodesource.com/setup_9.x | bash -
 RUN apt-get install -y nodejs
 RUN echo "banner Node-PI" >> $HOME/.bashrc
 RUN echo "git --version" >> $HOME/.bashrc
 RUN echo "echo -n 'node:' && node -v" >> $HOME/.bashrc
 RUN echo "echo -n 'npm:' && npm -v" >> $HOME/.bashrc

 RUN mkdir /workdir
 WORKDIR /workdir
 RUN git clone https://github.com/OlivierLD/node.pi.git
 WORKDIR /workdir/node.pi
 RUN npm install

 EXPOSE 9876
 CMD ["npm", "start"]

In this case, the full Docker image creation (named oliv-nodepi below) comes down to 1 line (the one in bold red):
 $ docker build -t oliv-nodepi .
Sending build context to Docker daemon  752.6kB
Step 1/20 : FROM resin/raspberrypi3-debian:latest
 ---> c542b8f7a388
Step 2/20 : MAINTAINER Olivier LeDiouris 
 ---> Using cache
 ---> b2ff0d7c489f
Step 3/20 : ADD nodepi.banner.sh /
 ---> 535733298dd1
Step 4/20 : RUN echo "alias ll='ls -lisah'" >> $HOME/.bashrc
 ---> Running in 09baf7261a55
Removing intermediate container 09baf7261a55
 ---> 71e1e4c95663
Step 5/20 : RUN apt-get update
 ---> Running in 5d817a941a14
Get:1 http://security.debian.org jessie/updates InRelease [94.4 kB]
Get:2 http://archive.raspbian.org jessie InRelease [14.9 kB]
Get:3 http://archive.raspberrypi.org jessie InRelease [22.9 kB]

...

npm notice created a lockfile as package-lock.json. You should commit this file.
added 166 packages in 81.166s
Removing intermediate container 13986530db28
 ---> 051eb94b8a3c
Step 19/20 : EXPOSE 9876
 ---> Running in 67b587845fe0
Removing intermediate container 67b587845fe0
 ---> 46973b7ba9ac
Step 20/20 : CMD ["npm", "start"]
 ---> Running in 153bf2ea02ad
Removing intermediate container 153bf2ea02ad
 ---> 6bf3d76d38ae
Successfully built 6bf3d76d38ae
Successfully tagged oliv-nodepi:latest
ed9a7d9042dddd3939b1788cf0e89d16f5273192a6456266507f072f90ce91bc
 $

Once the step above is completed, plug in your GPS, and run
 $ docker run -p 9876:9876 -t -i --privileged -v /dev/ttyUSB0:/dev/ttyUSB0 -d oliv-nodepi:latest
Then from a machine seeing the Raspberry PI on its network (it can be the Raspberry PI itself), reach http://raspi:9876/data/demos/gps.demo.wc.html in a browser.

This shows you the position the GPS has computed, and the satellites in sight.
You can also login to the image:
 $ docker run -it oliv-nodepi:latest /bin/bash

 #     #                                 ######    ###
 ##    #   ####   #####   ######         #     #    #
 # #   #  #    #  #    #  #              #     #    #
 #  #  #  #    #  #    #  #####   #####  ######     #
 #   # #  #    #  #    #  #              #          #
 #    ##  #    #  #    #  #              #          #
 #     #   ####   #####   ######         #         ###

 git version 2.1.4
 node:v9.11.1
 npm:5.6.0
 root@b9679d0d65a7:/workdir/node.pi#

... and do whatever you like.
The build operation needs to be done once.
There is no need to do it again as long as no change in the image is required.

Quick comment
So, with Docker, you do not deliver a software, you actually deliver an image (a virtual machine), on which a software is running.
This is indeed redefining the concept of portability that made Java and other JVM-aware languages so successful.
This may very well explain the rise of languages like Golang (aka Go).
It runs on my machine? Well, here is my machine! You can download and run it. Enjoy!

Saturday, March 17, 2018

Java Weather Station

This is a SwitchDocLabs SDLWeather80422 Weather Station, installed on the roof.

It is connected to a Raspberry PI A+, all the software is written in Java, no Python, no Arduino-like code, no C++.

There is an optional nodeJS server that runs on the Raspberry PI too, to enable WebSockets.

Find the core code here, and the example implementation here.

MySQL, PHP, Web Interface
Web-Components interface, pings the server every second.
WebSocket Web Interface, updated in real-time.
Some papers by John Shovic turned out to be very useful, specially in understanding what this debounce aspect is all about.

See it live here.

It even comes with a pebble application.

Sunday, September 10, 2017

Moving the Raspberry PI away from Swing

Rationale

This starts from a simple observation. A Raspberry PI can run on a boat, and consumes a very small amount of energy. It can do a lot of computations, logging, and multiplexing, among many others. It can run 24x7, without you noticing. It makes no noise, almost no light, and requires ridiculous amount of energy to run. Even a Raspberry PI Zero does this kind of job (for even less power), successfully.
One thing it is not good at is graphical UI. A graphical desktop is often too demanding on a small board like the Raspberry PI Zero. It becomes some times really slow, and cumbersome.
Running on it a program like OpenCPN seems absurd to me. Such a program runs fine on a bigger device, with several gigabytes of RAM available.
But, running a laptop 24x7 would be in many cases too demanding, specially on a sailboat, where everyone hates to run the engine ;)
I observed that at sea, I spend only a couple hours a day in front of the laptop, but it is often running by itself, doing some logging or calculations.
This is where it comes together, you could have a Raspberry PI Zero doing logging, multiplexing and what not, broadcasting require data on its own network (see the NMEA Multiplexer about that), then you would use a laptop whenever necessary, connecting on the Raspberry PI's network to get NMEA Data and more.
In addition, you can also use tablets and smart-phones, those devices know how to connect to a network, and have great rendering capabilities.
A problem is that writing a native application on those devices requires specific knowledge of the operating system, those skills are often redundant. iOS, Android, JavaFx, Swing all have UI rendering capabilities, but they're all totally different, and the learning curve for each of them is not always smooth.
A solution would be to write the UI part of the applications using HTML. Whatever OS runs on your laptop, tablet or smartphone (Windows, MacOS, iOS, Linux, Android, etc), you have a browser available, supporting HTML5 (if it does not, you should really upgrade it).
HTML5 and JavaScript have been gaining a lot of momentum in the recent years, new frameworks like jQuery, ionic, ReactJS, ...) appear every day, and provide really rich and nice UI.
My feeling would be to go down this route whenever possible, that would save a lot of efforts, and provide a pretty cool Graphical User Interface (GUI). I have written a lot of GUI in Swing. It would be now time to upgrade it. Re-writing them using JavaFX does not sound like the right choice. If I have to learn a new language to build a modern GUI, for now I'd rather use JavaScript and HTML5. This way, the same code runs whenever a browser exists... You have REST APIs available on the server (again, a Raspberry PI, even the Zero does the job well), and you use AJAX and Promises to get to them from the Web UI (WebSockets are also a realistic option, tested). The computation required to produce the payload returned by the REST services (often in json format) is easily supported by a Raspberry PI, and the complexity of the UI rendering is 100% taken care of by the browser, running on a more powerful device.

Implementation

To make sure all this is realistic, we have a REST implementation of a Tide Server, available here.
First, we have defined the REST Services, like
 /GET /tide-stations
 /GET /tide-stations/{station}
 /POST /tide-stations/{station}/wh?from=XXX&to=YYY
 /POST /tide-stations/{station}/wh/details?from=XXX&to=YYY
this is the easy part - and then an HTML5/JavaScript User Interface.


Harmonic coefficients are available for display


Period of time goes - in this UI - up to 1 month.


For one month, with harmonic coefficients, the volume of data transferred from the server is about 10Mb, it took about 14 seconds to get them.


In a most common case, it is around 25Kb.


This is running on a Raspberry PI, even a Raspberry PI Zero does the job without complaining.
There are a couple of challenges to address, JavaScript is not very well TimeZone equipped. But there are ways to get it to work.
That seems to be a viable approach.
Interestingly, even if we are trying here to address an energy problem - and not a budget one - a Raspberry PI Zero today cost just $10.

Saturday, December 31, 2016

NMEA Multiplexer, OpenCPN, GPSd...

I have been working on an NMEA Multiplexer that can run on small boards, like the Raspberry PI Zero. The code is available on github, see the documentation in the README.md.

It allows to mix all kinds of NMEA Sources into a single (or multiple) stream(s). You can read from Serial Ports, Log file(s), TCP, WebSocket, Sensors (like BME280, HTU21DF, LSM303, etc), merge those data and rebroadcast them on Serial port, TCP, Log file, WebSocket, GPSd, etc. UDP is being worked on.

Data can also be computed and injected in the output stream, like True Wind, Current direction and speed, etc.

As a graphical desktop can be cumbersome on small boards, the Multiplexer comes with a tiny HTTP server that provides a Web UI and REST services to allow remote Admin.

It also comes with several demos and samples
This all works just fine with OpenCPN, SeaWi, that can take TCP streams as NMEA Data Input.
I was also wondering about GPSd. I had some mixed feelings about it. Mostly, I was asking myself "Why should I parse GPSd json objects if I can parse NMEA Sentences?", and could not find any satisfying reason. The topic is mentioned on the GPSd web site's FAQ pages, but nothing clear (to me) came up from that.
Interestingly, OpenCPN can also take GPSd streams as input. But there is a trick.
The first GPSd exchange begins with a ?WATCH request. It is followed by a JSON Object like this:
  ?WATCH={"enable":true,"json":true}
... and here is the trick, OpenCPN sends a
 ?WATCH={"enable":true,"nmea":true}
This nmea option is "poorly" documented, but very useful. Instead of sending JSON objects, GPSd spits out the raw NMEA sentences, as they were read. Then GPSd is just a regular TCP stream, and OpenCPN already knows how to parse the NMEA sentences it delivers. This way, GPSd is not limited to strictly GPS-related sentences. It can convey all NMEA sentences, Boat Speed related, Wind related, etc. This is what the GPSd forwarder that comes with the Multiplexer is doing.
Happy Streaming, happy new year!

Sunday, July 24, 2016

Live Wallpaper, offline

The Live Wallpaper (in the navigation console) has been improved. Several bugs were fixed (like the one for the daylight), and you can now use it offline, like when no NMEA data are coming. This can be useful when you are not on the boat (like at home).

You can give a default position (in the preferences), and it will be used when no GPS Data is available.
New features:
  • The tilt is based on the Sun declination
  • You can have Moonlight and Sunlight
  • The tide curve of the closest station is displayed (closest in a radius of 100 nautical miles)
  • Sun and Moon positions are displayed (when the body is visible)
Code and (some) doc are available on GitHub.

Sunday, March 13, 2016

IoT for dummies

Internet of things, what's the big deal?
We have a device, that can read sensors on one side, and connect to the Internet on the other, nothing revolutionary here. The challenge is to reach the device, from somewhere on the Internet, the device that reads the sensors does not necessary have a public IP address.
This is why the key component is the IoT server.
The device (the one with sensors) pushes data on the IoT server. The server can then be read from anywhere on the net.

On top of that, you can push data to the IoT device. For example, in a room:
  • the device is connected to a temperature sensor, and pushes temperature data to the IoT server (and so, they can be read from a browser, smartphone...)
  • the device is connected to a relay, data can be sent to the device (like from a browser, smartphone...), to drive the relay. If the relay is connected to a heater, you can manage the room temperature.


Typically, you can have a Raspberry PI, Arduino, or similar board, reading - at home - a temperature sensor to publish the air temperature onto the IoT server. You access those data from a browser (from your office, your smart-phone, whatever). Then you may decide to turn the heater on at home, the Raspberry PI (or its friends) has a relay that drives the heater.

Many server also use some push technology (like WebSocket or similar), so a client can be notified. A server like Adafruit IO provides this service for free, Particle also does it to some extend.

In addition, Adafruit IO provides a REST interface, very slick.
See an example of a REST client - in Java - on github. It shows how to read and write data on the server.

A real IoT application is also featured here in github, along with its Adafruit IO dashboard. It reads a BME280 (for the air temperature), and provides a toggle button (switch) that drives a relay on the Raspberry PI, to turn a heater on or off.

Friday, December 11, 2015

Raspberry PI Zero is here

I just received mine. And it just works. It takes a ridiculous amount of energy, and works just like its bigger brothers. Just a couple of things to keep in mind:
  • Use the latest RasPian image, I used - successfully - the one from Nov-11, 2015 (2015-11-21-raspbian-jessie.img)
  • If you use a desktop USB keyboard, you need a powered USB hub
  • Once started, do not forget to expand your file system, so it uses all the space available on your SD card
I used it with the small Edimax Wireless USB dongle (the powered hub is not required for this one), I got started in no time. I was able to ssh to it and run everything on it! So far, it is an amazing machine. A real fully featured Linux computer for $5. And 20 times more memory than the one I had in my desk 25 years back...
. . .
That would remind many things to some of us... Amstrad, Amiga, Atari, wow! I must be getting old.

PS: About the pinout: Put the SD Card on top, the biggest Raspberry design under the board. The pin #1 (3V3) is at the top left of the header.

Thursday, November 19, 2015

Monitor the Boat, remotely

With a FONA connected on the Raspberry PI (where the NMEA Console is running), you can remotely monitor what's going on.
Sending a '?' returns the menu, all subsequent available commands are 1 or 2 character long.

The feature has been implemented - as usual - as a UserExit. This UserExit is in the Desktop repo, the FONA Java interface is here.
Add the following parameter to the command line:
 -ue:olivsoftdesktopuserexits.FONAUserExit

Just by sending an SMS, you can retrieve the data read by the Raspberry PI on board, like True Wind Speed, Battery Voltage, Air and Water Temperature, etc, all you need is a smart phone that can send and receive SMSs.

That sounds promising...

This assumes that the boat is docked in a place where there is SMS coverage, of course. I'm working on an Internet version, with a Particle Photon, or an ESP8266...

Saturday, June 06, 2015

Scala on the Raspberry PI

As Scala compiles its files as Java class-files, they run on a regular JVM, with a few Scala-specific jar-files in the classpath.
Even if it is not the most productive way to get work done, it is possible to compile Java files on the Raspberry PI, as well as Scala files. Personally, I prefer to develop in an IDE, and use FTP to push the classes to the Raspberry PI, it's much faster, and the IDE is much more productive than vi.
The explanations I found here got me started.
And again, as Scala runs on a Java Virtual Machine (JVM), all the work done with PI4J is fully available from Scala.
The following code (available on github) shows how to read a BMP180 from Scala:
 import adafruiti2c.sensor.AdafruitBMP180

 object Scala_101 {
   def main(args: Array[String]) {
     println("Hello, Scala world!")
     val bmp180 = new AdafruitBMP180
     try {
       val temp  = bmp180.readTemperature
       val press = bmp180.readPressure / 100
       println(s"CPU Temperature   :  ${SystemInfo.getCpuTemperature}\272C")
       println(s"Temp:${temp}\272C, Press:${press} hPa")
     } catch {
       case ex: Exception => {
         println(ex.toString())
       }
     }
   }
 }
  
To run it, I used a script like this:
 #!/bin/bash
 # 
 SCALA_HOME=/home/pi/.sbt/boot/scala-2.10.3
 PI4J_HOME=/opt/pi4j
 #
 CP=$SCALA_HOME/lib/scala-library.jar
 # 
 CP=$CP:$PI4J_HOME/lib/pi4j-core.jar
 CP=$CP:../AdafruitI2C/classes
 CP=$CP:./out/production/Scala.101
 # 
 sudo java -classpath "$CP" Scala_101
  
And all is good, the output is
 Hello,Scala world!
 CPU Temperature  :40.6°C
 Temp:22.5°C, Press:1010.73 hPa

Thursday, May 07, 2015

Web Console improvements

With the Raspberry PI now able to behave as an Access Point, almost all devices can connect to it (Apple, Android, all kind of phones or tablets). Using an HTML5 enabled browser makes live easier than writing a dedicated application for the given Operating System.
That's why we now have some improvements in the Web Console:
See above the "Night" scheme, with or without borders for the displays
Same for the "Day" scheme.

The HTML5 console is accessed from http://machine:port/html5/console.html, default port being 9999.
If you have installed node.js and the WebSocket user-exit, then you access the WebSocket console from http://machine:9876/data/console.ws.html
It is still self-contained, no external framework is used (like JQuery et al). Those are great - for sure - but this is to be run on a boat at sea, with Internet out of reach.
Notice on the snapshots that several data come from some sensors hooked-up on the Raspberry PI. They can be shown or hidden from the preferences or from the Console Admin page.
Notice that the Console Admin page has been removed from the Console, it can now be accessed from a separate URL, on the admin port (8080 by default) at http://machine:port/html5/admin.html, started when the console is in headless mode:
The default values on the pages above come from the Desktop Preferences.
And there is now a CLI (Command Line Interface) for to access those preferences from a non-graphical environment (like on the Raspberry PI).
The class to launch is olivsoftdesktop.PreferencesCLI. An entry will added soon in the User Interfaces.

All this runs fine on the Raspberry PI, all the snapshots above have been taken with the Raspberry PI run node.js as server.

An idea...

Anyone with a smartphone or a tablet can access those live data. The problem is to type in the right URL...
Once you have chosen your configuration (IP address and ports), you can generate QR Code (https://www.the-qrcode-generator.com/ worked for me), print them, and post them somewhere in the boat. Whoever wants to reach the data just uses his QR Scanner - all smart stuffs have at least one - and boom! You're in!

HTML5 console

HTML5 WebSocket console

HTML5 Admin console
Those pages support query string parameters theme and border. theme can be 'white' or 'black', border can be 'Y' or 'N'.
Like in http://machine:9999/html5/console.html?theme=white&border=N

There are several QR Code generators, including some you can run off-line. This one works just fine, and can be installed on the computer on the boat, so you can generate your codes from anywhere. As you can see here.

Thursday, March 05, 2015

OpenCPN on the Raspberry PI model 2

I just received the new Raspberry PI model B, it comes with 1Gb of RAM.
And it seems that OpenCPN runs just fine on it!
I started from the last NOOBS available from the Raspberry PI website.
Here are the steps I had to go through to build it:
Prompt> sudo apt-get update
Prompt> sudo apt-get install -y libgtk2.0-dev gettext git-core cmake gpsd gpsd-clients \
  libgps-dev build-essential wx-common libwxgtk2.8-dev \
  libglu1-mesa-dev libgtk2.0-dev wx2.8-headers \
  libbz2-dev libtinyxml-dev libsdl1.2debian xcalib

Prompt> git clone https://github.com/seandepagnier/OpenCPN.git
Prompt> cd OpenCPN/
Prompt> mkdir build
Prompt> cd build
Prompt> cmake ../
Prompt> make
Prompt> sudo make install
After that, at the prompt you enter:
Prompt> opencpn &
And that's it!
And by the way, it works the same on all the Ubuntu-like distributions I tested.

Thursday, January 08, 2015

Version 3.0.1.5 available

New version - for both the Weather Wizard and the Navigation Console - 3.0.1.5 is available, at the usual place.
Few new features, minor UI improvements, and big uptake for the Raspberry PI.
A new feature for the Weather Wizard in headless mode, you can download several composites at the same time (one after the other actually). Use the -composite: and -pattern: parameters, separate each value with a comma.
For example, on Windows:
 set PRMS=-composite:./patterns/01.Favorites/01.3.00.Pacific.Sfc.500.Tropic.GRIB.ptrn
 set PRMS=%PRMS%,./patterns/01.Favorites/06.01.AllPac.Faxes.Satellite.ptrn
 set PRMS=%PRMS% -interval:360 
 set PRMS=%PRMS% "-pattern:/yyyy/MM-MMM | | yyyy-MM-dd_HHmmss_z | _Pacific | waz,/yyyy/MM-MMM | | yyyy-MM-dd_HHmmss_z | _Pacific.SatPic | waz"
 ::
 set command=java %JAVA_OPTIONS% -client -classpath "%CP%" -Dheadless=true main.splash.Splasher %PRMS%
On Linux and Mac:
 PRMS=-composite:./patterns/01.Favorites/01.3.00.Pacific.Sfc.500.Tropic.GRIB.ptrn
 PRMS=$PRMS,./patterns/01.Favorites/06.01.AllPac.Faxes.Satellite.ptrn
 PRMS=$PRMS -interval:360 
 PRMS=$PRMS "-pattern:/yyyy/MM-MMM | | yyyy-MM-dd_HHmmss_z | _Pacific | waz,/yyyy/MM-MMM | | yyyy-MM-dd_HHmmss_z | _Pacific.SatPic | waz"
 #
 command=java $JAVA_OPTIONS -client -classpath "$CP" -Dheadless=true main.splash.Splasher $PRMS

Monday, July 28, 2014

The self-sufficient Raspberry PI

Here is a picture of the Raspberry PI setup to log the data from a GPS, and fueled by a solar panel.
This one has the small touchscreen from Adafruit (zoom on it). The tray and the breadboard are not mandatory in this setting. Next step will be to use two servos to orient the solar panel so it faces the sun. This would be based on the project available on Google Code. That's the cool thing about using Java on the Raspberry PI. Whatever works in Java works on it.

Thursday, July 10, 2014

How the Raspberry PI makes sense on board

A computer can be helpful on board, for several reasons.
  • Some useful navigation softwares need one to run (SailMail, for example, that allows you to receive faxes through the SSB, send and receive emails, all kinds of useful things)
  • Some chart plotter softwares (like OpenCPN, the best) require a computer to run as well, to plot the current position on an electronic chart.
  • The on-board electronics can be read from a computer (through NMEA, or some other proprietary protocols), and the data they emit can be used as parameters for calculations, like performance evaluation, routing, current estimation (tricky, but so useful).

Electricity can be precious on board (specially on a sail boat, where it is in short supply), it is required to run several important devices, like the autopilot, the water maker, and to some extend, the fridge. A laptop can draw a substantial amount of current, specially if its battery is old (2 to 3 amps and more, I have evidences). In those conditions, leaving it on all the time can be questionable. On top of that, turning it down, and turning it back up takes time...

Serial port access

The data we are interested in usually (like in 99% of the cases) come from a Serial port (USB, or 9-pins). A big detail to mention is that a Serial port can only be accessed by one process at a time. That means that when your chart plotter accesses the serial port, no other program can access it, even it is is not the same data the other program is interested in.
The chart plotter will be interested in the GPS data, another soft might be interested in the wind data; but there is no way around, one program, one port.
GPSd does not address this issue (even if it pretends to), and - in my opinion - makes things more complex. It's only interested in GPS data (not in Speed Through Water, not in Wind Data, etc), but it locks the port like everyone else, and just rebroadcasts them in another bloody format! Why isn't it just rebroadcasting the NMEA sentences as they've been read?.., I have no idea.

Introducing the Raspberry PI

The Raspberry PI does much more than the boards like Arduino, Sparkfun, Beaglebones, and others (those are great, don't get me wrong, I am not spitting in the soup), it is a fully featured Linux (Debian) computer, that can - as such - do all a computer can do, like multi tasking, multi threading, remote access (SSH, VNC), network access (Ethernet and Wireless). It only has 512 Mb of RAM. But that is enough, as we will see. Its hard disk is replaced by an SD card, ranging from 4GB to whatever you want. I use 16Gb cards. And at work, it draws less than 500mA, which is ridiculous. You can plug several kinds of screen on the Raspberry PI, a TV screen using an HDMI port, or a rear camera car monitor (3.5", 4.3", or 7") using an RCA port. Turning those screens off when not needed will also contribute to save some energy.
The Raspberry PI can read the serial port - Ok, exclusively - and re-broadcast the data on whatever channel (TCP, HTTP, UDP, RMI, whatever). This way they can be accessed from this channel, the data remaining the same. As soon as a device is turned on, it can join the ad-hoc network created by the Raspberry PI, and immediately read the data it broadcasts.

A program like OpenCPN is smart enough to support several kinds of channels.
HTTP is also an option to consider, smart phones and tablets are ready for that, without any modifications. HTML5 and CCS3 will do the job. Those devices have browsers that understand those technologies. It is very easy to display the data read by the Raspberry PI on an iPhone.
In addition, the Raspberry PI can log the data read from the NMEA Port, its SD card is big enough to log several days of data. Another feature of the Raspberry PI is its GPIO Header. This is the bridge to the world of sensors. For example, the BMP180 will allow the Raspberry PI to read the air temperature and the atmospheric pressure. Those data can very well be injected in the NMEA stream (they both have an NMEA equivalent). Some navigation station already provide this kind of interfaces, but in case yours does not, you will then get those data for less than $10.
See some implementation details here.

Even further, it is not difficult to come up with a small setting the Raspberry PI can use to monitor the tension of the batteries on the boat. This one does not - as far as I know - have an NMEA equivalent. It can be injected in the NMEA stream though, it will be considered as a custom sentence. And it can be logged too, along with the rest.
If needed, the data can be displayed on the Raspberry PI as they are read. I did some tests with a graphical interface, as well as in character mode, just to keep the energy consumption as low as possible. The screen is turned off when not needed, for the same reason.

Useful links



Basically, you can reproduce the settings used during the last Aremica's Cup, for a tiny fraction of their budget! The Raspberry PI is less than $40.
Compare with this...

If it does not work for you, do let me know, it should. I might be able to help.

Friday, May 09, 2014

Embedded small screen on the Raspberry PI

Adafruit provides a small touch screen, the size of the Raspberry PI... It works like a charm! See below...
It comes with detailed explanations about the way to setup the software.
The only thing I had to mess with was the hardware..., the way to hook up the screen on the board.
I must be a software guy... ;0)
And with the 26-pin connector on the right (under the board), you can still use the cobbler.
The ncurses menu at startup turned out to be useful.
There is a soft keyboard that can be useful too in this context. Install it by running
 sudo apt-get install matchbox-keyboard
Here is what it can look like:

The screenshot above is taken using scrot,
 sudo apt-get install scrot
Cool stuff.
A Swing UI works fine on this little screen, even if it may need to be reworked to fit the screen. But it's all good.