• pilotpi-rpios
  • PilotPi with Raspberry Pi OS
• Developer Quick Start
  • OS Image
  • Setting up Access (Optional)
    • Hostname and mDNS
  • Setting up OS
    • config.txt
    • cmdline.txt
    • rc.local
    • CSI camera
  • Building the code
    • Cross build for Raspberry Pi OS
  • Alternative build method (using docker)
  • Post-configuration
    • Mixer file
    • External compass
• Raspberry Pi 2/3 Navio2 Autopilot
  • OS Image
  • Setting up Access
• Expand the Filesystem
• Disable Navio RGB Overlay
• Changing Hostnames
• Setting up Avahi (Zeroconf)
• PilotPi with Ubuntu Server
  • Developer Quick Start
    • OS Image
      • armhf
      • arm64
      • Latest OS
    • First boot
      • Before booting
      • WiFi region
      • Hostname and mDNS
      • Password-less Auth (Optional)
    • Setting up OS
      • config.txt
      • cmdline.txt
      • rc.local
      • CSI camera
    • Building the code
      • Set RPi upload target
      • Build for armhf target
      • Alternative build method for armhf (using docker)
      • Build for arm64 target
      • Alternative build method for arm64 (using docker)
      • Manually run PX4
    • Post-configuration
      • Hostname and mDNS
      • Password-less Auth (Optional)
    • Setting up OS
      • config.txt
      • cmdline.txt
      • rc.local
      • CSI camera
    • Building the code
      • Set RPi upload target
      • Build for armhf target
      • Alternative build method for armhf (using docker)
      • Build for arm64 target
      • Alternative build method for arm64 (using docker)
      • Manually run PX4
    • Post-configuration
  • Configuring a SSH Public-Key
  • Testing file transfer
  • Building the Code
    • Cross-compiler Build
    • Native Build
  • Autostart

pilotpi-rpios

PilotPi with Raspberry Pi OS

Developer Quick Start

OS Image

The latest official Raspberry Pi OS Lite image is always recommended.

To install you must already have a working SSH connection to RPi.

Setting up Access (Optional)

Hostname and mDNS

mDNS helps you connect to your RasPi with hostname instead of IP address.

sudo raspi-config

Navigate to Network Options > Hostname. Set and exit. You may want to setup passwordless auth as well.

Setting up OS

config.txt

sudo nano /boot/config.txt

Replace the file with:

# enable sc16is752 overlay
dtoverlay=sc16is752-spi1
# enable I2C-1 and set the frequency to 400KHz
dtparam=i2#c_arm=on,i2c_arm_baudrate=400000
# enable spidev0.0
dtparam=spi=on
# enable RC input
enable_uart=1
# enable I2C-0
dtparam=i2c_vc=on
# switch Bluetooth to miniuart
dtoverlay=miniuart-bt

cmdline.txt

sudo raspi-config

Interfacing Options > Serial > login shell = No > hardware = Yes. Enable UART but without a login shell on it.

sudo nano /boot/cmdline.txt
```#

Append `isolcpus=2` after the last word.
The whole file would be:

```sh
console=tty1 root=PARTUUID=xxxxxxxx-xx rootfstype=ext4 elevator=deadline fsck.repair=yes rootwait isolcpus=2

This tells the Linux kernel not to schedule any process on CPU core 2. We will manually run PX4 onto that core later.

Reboot and SSH onto your RasPi.

Check UART interface:

ls /dev/tty*

There should be /dev/ttyAMA0, /dev/ttySC0 and /dev/ttySC1.

Check I2C interface:

ls /dev/i2c*

There should be /dev/i2c-0 and /dev/i2c-1

Check SPI interface

ls /dev/spidev*

There should be /dev/spidev0.0.

rc.local

In this section we will configure the auto-start script in rc.local.

sudo nano /etc/rc.local

Append below content to the file above exit 0:

echo "25" > /sys/class/gpio/export
echo "in" > /sys/class/gpio/gpio25/direction
if [ $(cat /sys/class/gpio/gpio25/value) -eq 1 ] ; then
        echo "Launching PX4"
        cd /home/pi/px4 ; nohup taskset -c 2 ./bin/px4 -d -s pilotpi_mc.config 2 &> 1 > /home/pi/px4/px4.log &
fi
echo "25" > /sys/class/gpio/unexport

Save and exit.

:::note Don’t forget to turn off the switch when it is not needed. :::

CSI camera

:::note Enable CSI camera will stop anything works on I2C-0. :::

sudo raspi-config

Interfacing Options > Camera

Building the code

To get the very latest version onto your computer, enter the following command into a terminal:

git clone https://github.com/PX4/PX4-Autopilot.git --recursive

:::note This is all you need to do just to build the latest code. :::

Cross build for Raspberry Pi OS

Set the IP (or hostname) of your RPi using:

export AUTOPILOT_HOST=192.168.X.X

or

export AUTOPILOT_HOST=pi_hostname.local

Build the executable file:

cd PX4-Autopilot
make scumaker_pilotpi_default

Then upload it with:

make scumaker_pilotpi_default upload

Connect over ssh and run it with:

cd px4
sudo taskset -c 2 ./bin/px4 -s pilotpi_mc.config

Now PX4 is started with multi-rotor configuration.

If you encountered the similar problem executing bin/px4 on your Pi as following:

bin/px4: /lib/xxxx/xxxx: version `GLIBC_2.29' not found (required by bin/px4)

Then you should compile with docker instead.

Before proceeding to next step, clear the existing building at first:

rm -rf build/scumaker_pilotpi_default

Alternative build method (using docker)

The following method can provide the same tool-sets deployed in CI.

If you are compiling for the first time with docker, please refer to the offical docs.

Execute the command in PX4-Autopilot folder:

./Tools/docker_run.sh "export AUTOPILOT_HOST=192.168.X.X; export NO_NINJA_BUILD=1; make scumaker_pilotpi_default upload"

:::note mDNS is not supported within docker. You must specify the correct IP address every time when uploading. :::

:::note If your IDE doesn’t support ninja build, NO_NINJA_BUILD=1 option will help. You can compile without uploading too. Just remove upload target. :::

It is also possible to just compile the code with command:

./Tools/docker_run.sh "make scumaker_pilotpi_default"

Post-configuration

You need to check these extra items to get your vehicle work properly.

Mixer file

Mixer file is defined in pilotpi_xx.conf:

mixer load /dev/pwm_output0 etc/mixers/quad_x.main.mix

All available mixers are stored in etc/mixers. You can create one by yourself as well.

External compass

In the startup script(*.config), you will find

# external GPS & compass
gps start -d /dev/ttySC0 -i uart -p ubx -s
#hmc5883 start -X
#ist8310 start -X

Uncomment the correct one for your case. Not sure which compass comes up with your GPS module? Execute the following commands and see the output:

sudo apt-get update
sudo apt-get install i2c-tools
i2cdetect -y 0

Sample output:

     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:          -- -- -- -- -- -- -- -- -- -- -- 0e -- 
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- 1e -- 
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- 
70: -- -- -- -- -- -- -- --

1e indicates a HMC5883 based compass is mounted on external I2C bus. Similarly, IST8310 has a value of 0e.

:::note Generally you only have one of them. Other devices will also be displayed here if they are connected to external I2C bus.(/dev/i2c-0) :::

Raspberry Pi 2/3 Navio2 Autopilot

:::warning PX4 does not manufacture this (or any) autopilot. Contact the manufacturer for hardware support or compliance issues. :::

:::warning PX4 support for this flight controller is experimental. :::

This is the developer “quickstart” for Raspberry Pi 2/3 Navio2 autopilots. It allows you to build PX4 and transfer to the RasPi, or build natively.

OS Image

Use the Emlid RT Raspbian image for Navio 2. The default image will have most of the setup procedures shown below already done.

:::warning Make sure not to upgrade the system (more specifically the kernel). By upgrading, a new kernel can get installed which lacks the necessary HW support (you can check with ls /sys/class/pwm, the directory should not be empty). :::

Setting up Access

The Raspbian image has SSH setup already. Username is “pi” and password is “raspberry”. You can connect to your RPi2/3 over a network (Ethernet is set to come up with DHCP by default) and then proceed to configure WiFi access. We assume that the username and password remain at their defaults for the purpose of this guide.

To setup the RPi2/3 to join your local wifi, follow this guide.

Find the IP address of your Pi from your network, and then you can proceed to connect to it using SSH.

ssh pi@<IP-ADDRESS>

Expand the Filesystem

After installing the OS and connecting to it, make sure to expand the Filesystem, so there is enough space on the SD Card.

Disable Navio RGB Overlay

The existing Navio RGB overlay claims GPIOs used by PX4 for RGB Led. Edit /boot/config.txt by commenting the line enabling the navio-rgb overlay.

#dtoverlay=navio-rgb

Changing Hostnames

To avoid conflicts with any other RPis on the network, we advise you to change the default hostname to something sensible. We used “px4autopilot” for our setup. Connect to the Pi via SSH and follow the below instructions.

Edit the hostname file:

sudo nano /etc/hostname

Change raspberry to whatever hostname you want (one word with limited characters apply)

Next you need to change the hosts file:

sudo nano /etc/hosts

Change the entry 127.0.1.1 raspberry to 127.0.1.1 <YOURNEWHOSTNAME>

Reboot the Pi after this step is completed to allow it to re-associate with your network.

Setting up Avahi (Zeroconf)

To make connecting to the Pi easier, we recommend setting up Avahi (Zeroconf) which allows easy access to the Pi from any network by directly specifying its hostname.

PilotPi with Ubuntu Server

:::warning Ubuntu Server on RPi 4B consumes a lot of current and generates a lot of heat. Design for better heat dissipation and high power consumption when using this hardware. :::

Developer Quick Start

OS Image

Both armhf and arm64 arch are supported.

armhf

• Ubuntu Server 18.04.5 for RPi2
• Ubuntu Server 18.04.5 for RPi3
• Ubuntu Server 18.04.5 for RPi4
• Ubuntu Server 20.04.1 for RPi 2/3/4

arm64

• Ubuntu Server 18.04.5 for RPi3
• Ubuntu Server 18.04.5 for RPi4
• Ubuntu Server 20.04.1 for RPi 3/4

Latest OS

Please refer to official cdimage page for any new updates.

First boot

When setting up RaPi’s WiFi for the first time we recommended using a wired Ethernet connection between your home router and RPi, and a monitor and keyboard.

Before booting

Mount the SD card onto your computer and modify the network settings. Please follow the official instruction here.

Now plug the SD card onto your Pi and boot for the first time. Make sure you have shell access to the RPi - either SSH connection over wired Ethernet, or direct accessing with keyboard and monitor.

WiFi region

First install required package:

sudo apt-get install crda

Edit the file /etc/default/crda to change the correct WiFi region. Reference List

sudo nano /etc/default/crda

Then your Pi will able to join your WiFi network after reboot.

Hostname and mDNS

Let’s set up hostname at first.

sudo nano /etc/hostname

Change the hostname to whatever you like. Then install the package required by mDNS:

sudo apt-get update
sudo apt-get install avahi-daemon

Perform a reboot. raspi

sudo reboot

Regain the accessibility through WiFi connection after the above operation.

ssh ubuntu@pi_hostname.local

Password-less Auth (Optional)

You may want to setup passwordless auth as well.

Setting up OS

config.txt

The corresponding file in Ubuntu is /boot/firmware/usercfg.txt.

sudo nano /boot/firmware/usercfg.txt

Replace the file with:

# enable sc16is752 overlay
dtoverlay=sc16is752-spi1
# enable I2C-1 and set the frequency to 400KHz
dtparam=i2c_arm=on,i2c_arm_baudrate=400000
# enable spidev0.0
dtparam=spi=on
# enable RC input
enable_uart=1
# enable I2C-0
dtparam=i2c_vc=on
# switch Bluetooth to miniuart
dtoverlay=miniuart-bt

cmdline.txt

On Ubuntu Server 20.04:

sudo nano /boot/firmware/cmdline.txt

On Ubuntu Server 18.04 or earlier, nobtcmd.txt and btcmd.txt should both be modified.

sudo nano /boot/firmware/nobtcmd.txt

Find console=/dev/ttyAMA0,115200 and remove that part to disable the login shell on serial interface.

Append isolcpus=2 after the last word. The whole file will then look like:

net.ifnames=0 dwc_otg.lpm_enable=0 console=tty1 root=LABEL=writable rootfstype=ext4 elevator=deadline rootwait fixrtc isolcpus=2

The above line tells the Linux kernel do not schedule any process on CPU core 2. We will manually run PX4 onto that core later.

Reboot and SSH onto your Pi.

Check UART interface:

ls /dev/tty*

There should be /dev/ttyAMA0, /dev/ttySC0 and /dev/ttySC1.

Check I2C interface:

ls /dev/i2c*

There should be /dev/i2c-0 and /dev/i2c-1

Check SPI interface:

ls /dev/spidev*

There should be /dev/spidev0.0.

rc.local

In this section we will configure the auto-start script in rc.local. Note that we need to create this file, as it is not present on a fresh Ubuntu OS.

sudo nano /etc/rc.local

Append the content below to the file:

#!/bin/sh

echo "25" > /sys/class/gpio/export
echo "in" > /sys/class/gpio/gpio25/direction
if [ $(cat /sys/class/gpio/gpio25/value) -eq 1 ] ; then
        echo "Launching PX4"
        cd /home/ubuntu/px4 ; nohup taskset -c 2 ./bin/px4 -d -s pilotpi_mc.config 2 &> 1 >raspi
echo "25" > /sys/class/gpio/unexport

exit 0

Save and exit. Then set the correct permissions:

sudo chmod +x /etc/rc.local

:::note Don’t forget to turn off the switch when it is not needed! :::

CSI camera

:::warning Enable CSI camera will stop anything works on I2C-0. :::

sudo nano /boot/firmware/usercfg.txt

Append the following line at the end of file:

start_x=1

Building the code

To get the very latest version onto your computer, enter the following command into a terminal:

git clone https://github.com/PX4/PX4-Autopilot.git --recursive

:::note This is all you need to do just to build the latest code. :::

Set RPi upload target

Set the IP (or hostname) of your RPi using:

export AUTOPILOT_HOST=192.168.X.X

or

export AUTOPILOT_HOST=pi_hostname.local

Additionally, we need to set the username:

export AUTOPILOT_USER=ubuntu

Build for armhf target

Build the executable file:

cd Firmware
make scumaker_pilotpi_default

Then upload it with:

make scumaker_pilotpi_default upload

Alternative build method for armhf (using docker)

If you are compiling for the first time with docker, please refer to the offical docs.

Execute the command in firmware folder:

./Tools/docker_run.sh "export AUTOPILOT_HOST=192.168.X.X; export AUTOPILOT_USER=ubuntu; export NO_NINJA_BUILD=1; make scumaker_pilotpi_default upload"

:::note mDNS is not supported within docker. You must specify the correct IP address every time when uploading. :::

:::note If your IDE doesn’t support ninja build, NO_NINJA_BUILD=1 option will help. You can compile without uploading too. Just remove upload target. :::

It is also possible to just compile the code with command:

./Tools/docker_run.sh "make scumaker_pilotpi_default"

Build for arm64 target

:::note This step requires aarch64-linux-gnu tool-chain to be installed. :::

Build the executable file:

cd PX4-Autopilot
make scumaker_pilotpi_arm64

Then upload it with:

make scumaker_pilotpi_arm64 upload

Alternative build method for arm64 (using docker)

If you are compiling for the first time with docker, please refer to the offical docs.

Execute the command in PX4-Autopilot folder:

./Tools/docker_run.sh "export AUTOPILOT_HOST=192.168.X.X; export AUTOPILOT_USER=ubuntu; export NO_NINJA_BUILD=1; make scumaker_pilotpi_arm64 upload"

:::note mDNS is not supported within docker. You must specify the correct IP address everytime when uploading. :::

:::note If your IDE doesn’t support ninja build, NO_NINJA_BUILD=1 option will help. You can compile without uploading too - just remove the upload target. :::

It is also possible to just compile the code with command:

./Tools/docker_run.sh "make scumaker_pilotpi_arm64"

Manually run PX4

Connect over SSH and run it with:

cd px4
sudo taskset -c 2 ./bin/px4 -s pilotpi_mc.config

Now PX4 is started with multi-rotor configuration.

If you encountered the similar problem executing bin/px4 on your Pi as following:

bin/px4: /lib/xxxx/xxxx: version `GLIBC_2.29' not found (required by bin/px4)

Then you should compile with docker instead.

Before proceeding to next step, clear the existing building at first: raspi

rm -rf build/scumaker_pilotpi_*

Then go back to the corresponding chapter above.

Post-configuration

Please refer to the instructions here

sudo apt-get install avahi-daemon
sudo insserv avahi-daemon

Next, setup the Avahi configuration file

sudo nano /etc/avahi/services/multiple.service

Add this to the file :

<?xml version="1.0" standalone='no'?>
<!DOCTYPE service-group SYSTEM "avahi-service.dtd">
<service-group># PilotPi with Ubuntu Server

:::warning
Ubuntu Server on RPi 4B consumes a lot of current and generates a lot of heat.
Design for better heat dissipation and high power consumption when using this hardware.
:::
<!--more-->

## Developer Quick Start

### OS Image

Both armhf and arm64 arch are supported.

#### armhf

- [Ubuntu Server 18.04.5 for RPi2](https://cdimage.ubuntu.com/releases/18.04.5/release/ubuntu-18.04.5-preinstalled-server-armhf+raspi2.img.xz)
- [Ubuntu Server 18.04.5 for RPi3](https://cdimage.ubuntu.com/releases/18.04.5/release/ubuntu-18.04.5-preinstalled-server-armhf+raspi3.img.xz.zsync)
- [Ubuntu Server 18.04.5 for RPi4](https://cdimage.ubuntu.com/releases/18.04.5/release/ubuntu-18.04.5-preinstalled-server-armhf+raspi4.img.xz)
- [Ubuntu Server 20.04.1 for RPi 2/3/4](https://cdimage.ubuntu.com/releases/20.04.1/release/ubuntu-20.04.2-preinstalled-server-arm64+raspi.img.xz)

#### arm64

- [Ubuntu Server 18.04.5 for RPi3](https://cdimage.ubuntu.com/releases/18.04.5/release/ubuntu-18.04.5-preinstalled-server-arm64+raspi3.img.xz)
- [Ubuntu Server 18.04.5 for RPi4](https://cdimage.ubuntu.com/releases/18.04.5/release/ubuntu-18.04.5-preinstalled-server-arm64+raspi4.img.xz)
- [Ubuntu Server 20.04.1 for RPi 3/4](https://cdimage.ubuntu.com/releases/20.04.1/release/ubuntu-20.04.2-preinstalled-server-arm64+raspi.img.xz)

#### Latest OS

Please refer to official [cdimage](https://cdimage.ubuntu.com/releases/) page for any new updates.

### First boot

When setting up RaPi's WiFi for the first time we recommended using a wired Ethernet connection between your home router and RPi, and a monitor and keyboard.

#### Before booting

Mount the SD card onto your computer and modify the network settings.
Please follow the official instruction [here](https://ubuntu.com/tutorials/how-to-install-ubuntu-on-your-raspberry-pi#3-wifi-or-ethernet).

Now plug the SD card onto your Pi and boot for the first time.
Make sure you have shell access to the RPi - either SSH connection over wired Ethernet, or direct accessing with keyboard and monitor.

#### WiFi region

First install required package:

```sh
sudo apt-get install crda

Edit the file /etc/default/crda to change the correct WiFi region. Reference List

sudo nano /etc/default/crda

Then your Pi will able to join your WiFi network after reboot.

Hostname and mDNS

Let’s set up hostname at first.

sudo nano /etc/hostname

Change the hostname to whatever you like. Then install the package required by mDNS:

sudo apt-get update
sudo apt-get install avahi-daemon

Perform a reboot. raspi

sudo reboot

Regain the accessibility through WiFi connection after the above operation.

ssh ubuntu@pi_hostname.local

Password-less Auth (Optional)

You may want to setup passwordless auth as well.

Setting up OS

config.txt

The corresponding file in Ubuntu is /boot/firmware/usercfg.txt.

sudo nano /boot/firmware/usercfg.txt

Replace the file with:

# enable sc16is752 overlay
dtoverlay=sc16is752-spi1
# enable I2C-1 and set the frequency to 400KHz
dtparam=i2c_arm=on,i2c_arm_baudrate=400000
# enable spidev0.0
dtparam=spi=on
# enable RC input
enable_uart=1
# enable I2C-0
dtparam=i2c_vc=on
# switch Bluetooth to miniuart
dtoverlay=miniuart-bt

cmdline.txt

On Ubuntu Server 20.04:

sudo nano /boot/firmware/cmdline.txt

On Ubuntu Server 18.04 or earlier, nobtcmd.txt and btcmd.txt should both be modified.

sudo nano /boot/firmware/nobtcmd.txt

Find console=/dev/ttyAMA0,115200 and remove that part to disable the login shell on serial interface.

Append isolcpus=2 after the last word. The whole file will then look like:

net.ifnames=0 dwc_otg.lpm_enable=0 console=tty1 root=LABEL=writable rootfstype=ext4 elevator=deadline rootwait fixrtc isolcpus=2

The above line tells the Linux kernel do not schedule any process on CPU core 2. We will manually run PX4 onto that core later.

Reboot and SSH onto your Pi.

Check UART interface:

ls /dev/tty*

There should be /dev/ttyAMA0, /dev/ttySC0 and /dev/ttySC1.

Check I2C interface:

ls /dev/i2c*

There should be /dev/i2c-0 and /dev/i2c-1

Check SPI interface:

ls /dev/spidev*

There should be /dev/spidev0.0.

rc.local

In this section we will configure the auto-start script in rc.local. Note that we need to create this file, as it is not present on a fresh Ubuntu OS.

sudo nano /etc/rc.local

Append the content below to the file:

#!/bin/sh

echo "25" > /sys/class/gpio/export
echo "in" > /sys/class/gpio/gpio25/direction
if [ $(cat /sys/class/gpio/gpio25/value) -eq 1 ] ; then
        echo "Launching PX4"
        cd /home/ubuntu/px4 ; nohup taskset -c 2 ./bin/px4 -d -s pilotpi_mc.config 2 &> 1 >raspi
echo "25" > /sys/class/gpio/unexport

exit 0

Save and exit. Then set the correct permissions:

sudo chmod +x /etc/rc.local

:::note Don’t forget to turn off the switch when it is not needed! :::

CSI camera

:::warning Enable CSI camera will stop anything works on I2C-0. :::

sudo nano /boot/firmware/usercfg.txt

Append the following line at the end of file:

start_x=1

Building the code

To get the very latest version onto your computer, enter the following command into a terminal:

git clone https://github.com/PX4/PX4-Autopilot.git --recursive

:::note This is all you need to do just to build the latest code. :::

Set RPi upload target

Set the IP (or hostname) of your RPi using:

export AUTOPILOT_HOST=192.168.X.X

or

export AUTOPILOT_HOST=pi_hostname.local

Additionally, we need to set the username:

export AUTOPILOT_USER=ubuntu

Build for armhf target

Build the executable file:

cd Firmware
make scumaker_pilotpi_default

Then upload it with:

make scumaker_pilotpi_default upload

Alternative build method for armhf (using docker)

If you are compiling for the first time with docker, please refer to the offical docs.

Execute the command in firmware folder:

./Tools/docker_run.sh "export AUTOPILOT_HOST=192.168.X.X; export AUTOPILOT_USER=ubuntu; export NO_NINJA_BUILD=1; make scumaker_pilotpi_default upload"

:::note mDNS is not supported within docker. You must specify the correct IP address every time when uploading. :::

:::note If your IDE doesn’t support ninja build, NO_NINJA_BUILD=1 option will help. You can compile without uploading too. Just remove upload target. :::

It is also possible to just compile the code with command:

./Tools/docker_run.sh "make scumaker_pilotpi_default"

Build for arm64 target

:::note This step requires aarch64-linux-gnu tool-chain to be installed. :::

Build the executable file:

cd PX4-Autopilot
make scumaker_pilotpi_arm64

Then upload it with:

make scumaker_pilotpi_arm64 upload

Alternative build method for arm64 (using docker)

If you are compiling for the first time with docker, please refer to the offical docs.

Execute the command in PX4-Autopilot folder:

./Tools/docker_run.sh "export AUTOPILOT_HOST=192.168.X.X; export AUTOPILOT_USER=ubuntu; export NO_NINJA_BUILD=1; make scumaker_pilotpi_arm64 upload"

:::note mDNS is not supported within docker. You must specify the correct IP address everytime when uploading. :::

:::note If your IDE doesn’t support ninja build, NO_NINJA_BUILD=1 option will help. You can compile without uploading too - just remove the upload target. :::

It is also possible to just compile the code with command:

./Tools/docker_run.sh "make scumaker_pilotpi_arm64"

Manually run PX4

Connect over SSH and run it with:

cd px4
sudo taskset -c 2 ./bin/px4 -s pilotpi_mc.config

Now PX4 is started with multi-rotor configuration.

If you encountered the similar problem executing bin/px4 on your Pi as following:

bin/px4: /lib/xxxx/xxxx: version `GLIBC_2.29' not found (required by bin/px4)

Then you should compile with docker instead.

Before proceeding to next step, clear the existing building at first: raspi

rm -rf build/scumaker_pilotpi_*

Then go back to the corresponding chapter above.

Post-configuration

Please refer to the instructions here

    <name replace-wildcards="yes">%h</name>
    <service>
            <type>_device-info._tcp</type>
            <port>0</port>
            <txt-record>model=RackMac</txt-record>
    </service>
    <service>
            <type>_ssh._tcp</type>
            <port>22</port>
    </service>

</service-group>

Restart the daemon

```sh
sudo /etc/init.d/avahi-daemon restart

And that’s it. You should be able to access your Pi directly by its hostname from any computer on the network.

:::tip You might have to add .local to the hostname to discover it. :::

Configuring a SSH Public-Key

In order to allow the PX4 development environment to automatically push executables to your board, you need to configure passwordless access to the RPi. We use the public-key authentication method for this.

To generate new SSH keys enter the following commands (Choose a sensible hostname such as <YOURNANME>@<YOURDEVICE>. Here we have used pi@px4autopilot)

These commands need to be run on the HOST development computer!

ssh-keygen -t rsa -C pi@px4autopilot

Upon entering this command, you’ll be asked where to save the key. We suggest you save it in the default location ($HOME/.ssh/id_rsa) by just hitting Enter.

Now you should see the files id_rsa and id_rsa.pub in your .ssh directory in your home folder:

ls ~/.ssh
authorized_keys  id_rsa  id_rsa.pub  known_hosts

The id_rsa file is your private key. Keep this on the development computer. The id_rsa.pub file is your public key. This is what you put on the targets you want to connect to.

To copy your public key to your Raspberry Pi, use the following command to append the public key to your authorized_keys file on the Pi, sending it over SSH:

cat ~/.ssh/id_rsa.pub | ssh pi@px4autopilot 'cat >> .ssh/authorized_keys'

Note that this time you will have to authenticate with your password (“raspberry” by default).

Now try ssh pi@px4autopilot and you should connect without a password prompt.

If you see a message “Agent admitted failure to sign using the key.” then add your RSA or DSA identities to the authentication agent, ssh-agent and the execute the following command:

ssh-add

If this did not work, delete your keys with rm ~/.ssh/id* and follow the instructions again.

Testing file transfer

We use SCP to transfer files from the development computer to the target board over a network (WiFi or Ethernet).

To test your setup, try pushing a file from the development PC to the Pi over the network now. Make sure the Pi has network access, and you can SSH into it.

echo "Hello" > hello.txt
scp hello.txt pi@px4autopilot:/home/pi/
rm hello.txt

This should copy over a “hello.txt” file into the home folder of your RPi. Validate that the file was indeed copied, and you can proceed to the next step.

Building the Code

Either build the source code on your development computer (“cross-compiler” build) or build it on the RaPi (“native” build) as shown below.

Cross-compiler Build

First install the standard developer environment on your Ubunto development computer.

Set the IP (or hostname) of your RPi using:

export AUTOPILOT_HOST=192.168.X.X

or

export AUTOPILOT_HOST=pi_hostname.domain

:::note The value of the environment variable should be set before the build, or make upload will fail to find your RPi. :::

Build the executable file:

cd PX4-Autopilot
make emlid_navio2 # for cross-compiler build

The “px4” executable file is in the directory build/emlid_navio2_default/. Make sure you can connect to your RPi over ssh, see instructions how to access your RPi.

Then upload it with:

cd PX4-Autopilot
make emlid_navio2 upload # for cross-compiler build

Then, connect over ssh and run it with (as root):

cd ~/px4
sudo ./bin/px4 -s px4.config

Native Build

A native build is one that you run directly on the Pi (the other option is to run builds on a development computer which cross-compiles for the Pi, and pushes the PX4 executable binary directly to the Pi).

Run these commands on the Pi to setup the build system on the Pi.

sudo apt-get update
sudo apt-get install cmake python-empy

Clone the Firmware directly onto the Pi then build the native build target (emlid_navio2_native).

git clone https://github.com/PX4/PX4-Autopilot.git --recursive
cd PX4-Autopilot
make emlid_navio2_native

The “px4” executable file is in the directory build/emlid_navio2_native/. Run it directly with:

sudo ./build/emlid_navio2_native/px4 build/emlid_navio2_native/etc -s ./posix-configs/rpi/px4.config

A successful build followed by executing px4 will give you something like this:

______  __   __    ___
| ___ \ \ \ / /   /   |
| |_/ /  \ V /   / /| |
|  __/   /   \  / /_| |
| |     / /^\ \ \___  |
\_|     \/   \/     |_/

px4 starting.


pxh>

Autostart

To autostart px4, add the following to the file /etc/rc.local (adjust it accordingly if you use native build), right before the exit 0 line:

cd /home/pi && ./bin/px4 -d -s px4.config > px4.log

The following wiki, pages and posts are tagged with jdlab