Skip to content

Latest commit

 

History

History
204 lines (148 loc) · 5.56 KB

pi3b-setup-guide.md

File metadata and controls

204 lines (148 loc) · 5.56 KB
Raw

Creating a Developer Environment

Requirements

- Raspberry Pi Model 3B (will likely work with newer models, but hasn't been
  tested).
- 8GB (or larger) Micro SD card
- WiFi with Internet

Initial Setup

Install Raspberry Pi OS

  1. Download the Raspberry Pi Imager from here. Plugin in your SD card, and then start the imager software.

  2. Select "Choose OS" -> "Raspberry Pi OS (other)" -> "Raspberry Pi OS Lite (64 bit)".

  3. Select "Choose Storage" -> Your SD card.

  4. Select the Gear button on the bottom right corner to bring up the settings.

  5. Enable the checkbox for "Set hostname" and change the hostname to "weatherbot.local".

  6. Enable the checkbox for "Enable SSH", make sure "Use password authentication" is selected.

  7. Enable the checkbox for "Set username and password". Keep the username as "user", and enter a password.

  8. Enable the checkbox for "Configure wireless LAN" and enter the SSID and password for your WiFi network. Optionally, set your Locale settings.

  9. Select "Save", then select "Write".

  10. It will take a couple minutes to write the OS to the SD card. Once it's complete, a window will pop up saying "Write Successful" and that the SD card can be removed. Select continue, and eject the SD card.

Configure the OS

Insert the SD card and power on the Raspberry Pi. Wait for the Raspberry Pi to boot, then find its IP address through your router's interface.

Open a terminal, enter

ssh {USERNAME}@{PI_IP_ADDRESS}

and insert the password you configured earlier when prompted. It should pull up a message from Debian, and launch a bash shell.

Copy the Code to the Pi

First, download the code to your local computer.

Open a new terminal and clone the git repository

cd PARENT_DIRECTORY
git clone https://github.com/XXXX/weatherbot

Copy the code directory over to the Raspberry Pi

scp -r weatherbot {USERNAME}@{PI_IP_ADDRESS}:~/

(Optional) Configure what you want

The next step will remove the Pi's ability to access the internet, so you should download and configure what you want now while you still have access to the internet.

Run the auto-setup script

SSH into the pi and run the included setup script

ssh {USERNAME}@{PI_IP_ADDRESS}
cd ~/weatherbot
./setup.sh

This should update the system, install the necessary packages, and configure the system as an access point ready for development and integration with the weatherbot pico.

After several minutes, the system should reboot on its own, and a new wifi network with the SSID "Weatherbot" should be found. If this is the case, the setup script worked, and you're done!

If you'd rather do the process manually, the process is detailed below. You can also read the bash scripts. Otherwise, the setup is completed.

Manual process

Update the System

In the bash session, enter

sudo apt update && sudo apt upgrade -y

Install helper programs

Now is the best time to install software. Later on, the raspberry pi will lose access to the internet, which will make it much harder to install software. This command will install what is needed for the pico-sdk and development.

sudo apt install -y git screen build-essential cmake gcc-arm-none-eabi \
libnewlib-arm-none-eabi libstdc++-arm-none-eabi-newlib

(Optional) Install whatever software you anticipate needing to use. Personally, I install

sudo apt install -y tmux vim htop

(Optional) Use public key authentication.

Install arduino-cli

First, we'll make an install directory

mkdir -p ~/.local/bin

Add the install directory to your path

echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.bashrc
source ~/.bashrc

Then, we'll install the arduino-cli

curl -fsSL https://raw.githubusercontent.com/arduino/arduino-cli/master/install.sh | BINDIR=~/.local/bin sh

Ensure it's working by checking the version

arduino-cli version

Copy the code over

TODO: Create instructions on downloading the code

If downloaded on your host machine, copy it over using the following command

scp -r {DOWNLOAD_FOLDER} {USERNAME}@{PI_IP_ADDRESS}:~/

Download the required libraries

In the code directory on the Raspberry Pi, run

sh auto_download_required.sh

This will download all the required libraries used in development to your Raspberry Pi.

We can test that things function properly by compiling the code

./compile.sh {CODE_DIRECTORY}

This should return no errors. This can be confirmed by

echo $?

returning 0, not 1 or other number.

Networking Setup

This step will setup the Raspberry Pi as its own access point router, allowing communication between the pico and a client device through the Raspberry Pi. This will remove internet access from the Raspberry Pi, so do everything that you need the internet for before continuing.

Once you are sure you want to continue, enter the following:

sudo -Es
./{CODE_DIRECTORY}/access-point-setup.sh

The Raspberry Pi will begin setting up the access point, lose connection, finish setting things up, then shut down. Once fully off, turn the raspberry pi back on.

Once on, it should create a wifi access point with the SSID "Weatherbot{RANDOM 6 DIGIT CODE}". Connect to the wifi network.

You should now be able to reconnect to the pi using

ssh [email protected]

If you connect, the development environment was successful.