Software Installation

You can download a ready-to-use ISO image containing the operating system (Raspberry Pi OS) along with all necessary libraries and Python code pre-installed here.

Follow the instructions on this page if you prefer a manual installation. The process can be challenging, so proceed only if you have a basic understanding of Linux distributions and are comfortable using the terminal.

What You Need

  • Raspberry Pi 5 - 8GB

  • Fan

  • Micro SD 128GB

  • Keyboard

  • Mouse

  • Screen

  • Micro HDMI to HDMI cable

  • Power supply for Raspberry Pi

  • 2 Raspberry Pi Camera Module 3 Wide Noir

  • Ethernet cable (optional)

  • Raspberry Pi Dac Pro (optional)

  • 2 servo motors

  • RFID sensor

  • Weight sensor

  • Temperature sensor

Launch Raspberry Pi OS for the First Time

  1. Connect a keyboard, mouse, and screen to the Raspberry Pi.

  2. Connect the Ethernet cable if you are going to use an Ethernet connection.

  3. If your SD card has the OS preinstalled, jump to step 4.

  4. If your SD card is empty, follow these instructions to download the OS and copy it to the SD: Raspberry Pi OS

  5. Insert the SD card and start the Raspberry Pi.

  6. Select your country and keyboard language and choose English as the general language.

  7. Type raspberry as the username and choose the password you want.

  8. Select your Wi-Fi or Ethernet connection.

  9. Choose your preferred web browser (Chromium).

  10. Select “Yes” when asked to update software (if you have your internet connection ready).

Updating the System

  1. Make sure you have internet connection and then update the software. Detailed instructions.

sudo apt update
sudo apt full-upgrade

Installing Needed Libraries

  1. Install OpenCV (library to manage video):

sudo apt install -y python3-opencv
sudo apt install -y opencv-data

  1. Install sound libraries: TODO check if pulseaudio is really needed.

sudo apt-get install pulseaudio
sudo apt-get install libportaudio2

  1. Install PyQT5 multimedia

sudo apt install python3-pyqt5.qtmultimedia
sudo apt install libqt5multimedia5-plugins

  1. Install VS Code (optional):

sudo apt install code

Creating a Python Environment and Installing pip Libraries

  1. Create a .env environment that includes all the packages installed in the global Python environment:

python -m venv --system-site-packages ~/.env

  1. Activate it:

source ~/.env/bin/activate

  1. Install needed libraries with pip:

pip install python-dateutil
pip install setuptools_scm
pip install sounddevice
pip install python-telegram-bot
pip install scipy

Changing Preferences

  1. Enable auto-login by clicking on Preferences -> Raspberry Pi Configuration -> Auto login.

  2. Add CPU monitors to the toolbar: Right-click on the toolbar, select Add/Remove Panel Items, and click Add. Select CPU Temperature Monitor and CPU Usage monitor.

Screen Settings

  1. Run the raspi-config.

sudo raspi-config

  1. Go to Display Options and disable blanking.

  2. Go to Advanced Options and select X11 (instead of Wayland).

Then, you need to configure the system to recognize a screen even if none is physically connected, so the software renders properly when accessed remotely. If you want to use an additional screen to present stimuli in the behavioral box, you need to set the system to work with two screens.

  1. Modify the file: /boot/firmware/cmdline.txt

sudo nano /boot/firmware/cmdline.txt

  1. Add the following text: vc4.fvc4.force_hotplug=1 if you are using only one screen, or vc4.fvc4.force_hotplug=3 if you are using two screens.

Explanation of values: (1 makes the system operate as if a screen is connected to HDMI 1). (2 makes the system operate as if a screen is connected to HDMI 2). (3 makes the system operate as if screens are connected to both HDMI 1 and HDMI 2).

  1. Change the resolution of the screen (or screens) to 1280 x 720: Preferences -> Screen Configuration.

I2C Communication

The temperature sensor and the weight sensor are connected to the I2C pins of the Raspberry.

  1. Run the raspi-config.

sudo raspi-config

  1. Go to Interface Options and enable I2C.

Accessing Pins via Hardware (for Servos) and Using UART Pin for Communication

To access the pins via hardware (faster and less jittery than software access), and to allow the use of the UART Pin for communication with the sensors you need to change two settings:

  1. Edit the /boot/firmware/config.txt file to include the following:

[all]
dtoverlay=pwm-2chan,pin=12,pin2=13,func=4,func2=4
enable_uart=1

Udev Rules for Consistent USB Device Naming

When you connect your Bpod device to the Raspberry Pi, it’s assigned a file path in /dev/, typically named ttyACM0. However, this name may vary (ttyACM1, ttyACM2, etc.), especially if you have other USB devices connected. To ensure a consistent and recognizable name, you can create a symbolic link to your Bpod device.

  1. Navigate to the Udev rules folder where custom rules are stored:

cd /etc/udev/rules.d

  1. Create a new rule file named 99-usb.rules:

sudo nano 99-usb.rules

  1. In the new rule file, add the following line to create a symbolic link named Bpod for any device that matches the pattern ttyACM*. This will allow any device named ttyACM* (where * can be any number) to be consistently linked, provided it’s connected to a specific USB port (in this case, the port associated with KERNELS==3-1:1.0).

KERNEL=="ttyACM*",KERNELS=="3-1:1.0",SYMLINK+="Bpod"

Note: The USB port identifier (KERNELS==“3-1:1.0”) may vary depending on the physical USB port you are using. For reference, the 3-1:1.0 port on the Raspberry Pi is usually the USB port next to the Ethernet connection. To confirm the exact port identifier, you can use the command dmesg after connecting the device.

  1. Activate the new rule by triggering the Udev process:

sudo udevadm trigger

Now, whenever the Bpod device is connected to the specified USB port, it will consistently appear with the symbolic link /dev/Bpod, regardless of its dynamic ttyACM* designation.

Install Training Village

  1. Make sure you are in .env environment.

source ~/.env/bin/activate

  1. Clone the repository inside your /home/raspberry/ directory.

git clone https://github.com/BrainCircuitsBehaviorLab/village.git

  1. Navigate to folder village.

cd /home/raspberry/village

  1. Install the repository.

pip install -e .

  1. You’re ready to start the system by simply running the main.py file. python /home/raspberry/village/village/main.py

Create an Alias and Run the Training Village

  1. Edit the .bashrc file:

nano ~/.bashrc

  1. Add the following line to create an alias:

alias run_village='/home/raspberry/.env/bin/python /home/raspberry/village/village/main.py'

  1. Reload the .bahsrc:

source ~/.bashrc

  1. Run Training village:

run_village