HATv3 mixes PWM and GPIO motors

[Ezward]

dt-duckiebot-interface/packages/hat_driver/include/hat_driver/hat.py

Line 45 in 97373d8

class HATv3(HATv2):

class HATv3(HATv2):
    _MOTOR_NUM_TO_PINS: Dict[int, MotorPins] = {
        1: MotorPins(10, 9, 8, MotorDirectionControl.PWM),
        2: MotorPins(33, 31, 13, MotorDirectionControl.GPIO)
    }

It appears on motor is controlled via the PCA9685 and one via GPIO. Given that DB21M has 5 RGB LEDs driven by PCA9685, it seem there would be conflicts in addressing the motor and LEDs. So that motor configuration seems wrong.

[tanij]

The DB21M has 4 LEDs, 2 in the front and 2 in the back. I believe the configuration is correct (although unusual).

[Ezward]

This is what I figured out from the code; the pwm value (the third value in the triplet), is always used as the channel on the PCA9685, so in the HATv3 code, motor 1 uses channel 8 on the PCA9685 and motor 2 used channel 13. The difference between the motors is how the two motor direction signals are sent to the motor controller; either via the PCA9685 for MotorDirectionControl.PWM or via a GPIO pin for MotorControlDirection.GPIO. I'm not sure why it was chosen to use GPIO for motor 2 but not motor 1, but I can see now how it works.

class HATv3(HATv2):
    _MOTOR_NUM_TO_PINS: Dict[int, MotorPins] = {
        1: MotorPins(10, 9, 8, MotorDirectionControl.PWM),
        2: MotorPins(33, 31, 13, MotorDirectionControl.GPIO)
    }

So here for motor 1 the first two values of the triplet, 10 and 9 refer to channels/pins on the PCA9685. For motor 2 they represent pin number of the Jetson Nano header.

It's not technically a bug, but it certainly is a little strange and confusing to someone (at least to me) unless they really dig into the code.