🛠 Pont série SoC ↔ MCU : le tunnel x86 vers ATmega32U4

Objectives

Set up the central pattern of the LattePanda Alpha: the serial bridge between the x86 PC (Core m3-7Y30) and its own ATmega32U4 coprocessor, connected via an internal USB invisible from the outside. The MCU sends measurement frames over Serial, and a Python script on the host reads them line by line. This is exactly the scheme of DFRobot's official "python_arduino_bridge" tutorial, without Firmata.

Hardware

• LattePanda Alpha (Windows or Linux) + compliant 12 V/3 A power supply
• Arduino IDE (on the MCU side) and Python 3 with pyserial on the host side: pip install pyserial
• On Linux: add the user to the dialout group (sudo usermod -aG dialout $USER, then log out and back in)
• Optionally, a potentiometer between 5 V, A0, and GND to see values change (implicit 5 V ADC reference for Leonardo)

Complete Code

On the firmware side (Arduino IDE, board "Arduino Leonardo"):

// stream_analog.ino — envoie A0 au PC hôte à 10 Hz
void setup { Serial.begin(9600); }
void loop {
  Serial.println(analogRead(A0));   // valeur ADC 10 bits : 0..1023
  delay(100);
}

On the host side (on the Alpha itself, Windows or any Linux distribution):

# host_reader.py — pas besoin de Firmata, tramage série simple
import serial  # pip install pyserial

# Windows : 'COM5'  |  Linux : '/dev/ttyACM0' (utilisateur dans le groupe 'dialout')
with serial.Serial('COM5', 9600, timeout=1) as ser:
    for line in ser:
        try:
            raw = int(line.strip)
            print(f'ADC={raw}  ({raw * 5.0 / 1023:.2f} V en supposant une référence 5 V)')
        except ValueError:
            pass  # ignorer les lignes partielles au démarrage

Expected Verification

The host terminal displays approximately 10 lines per second: ADC=0 (0.00 V) at rest, and values rising towards 1023 if a potentiometer is wired to A0. Ctrl+C stops the script without touching the firmware, which continues to transmit autonomously.

Common Errors

• Wrong port: on Windows, check the COM number in Device Manager; on Linux, ls /dev/ttyACM*.
• Incomplete lines on the first byte after opening the port: the try/except ValueError is essential.
• Two programs connected to the same port (IDE serial monitor open in parallel with the script): the script receives nothing.
• Forgetting the delay(100) on the MCU side: the host is flooded and the display becomes unreadable.