Self-Balancing Robot with a Pico and a PID Controller

Balancing is the classic control-theory challenge. The IMU measures tilt, a complementary filter cleans the signal, and a PID loop drives the wheels under the robot's center of mass.

Mission steps

1. Build a tall frame

   Put the battery high — a higher center of mass is actually easier to balance. Mount the IMU rigidly at the axle line.

2. Wire it

   IMU on I2C to the Pico, motor driver on PWM pins, battery to VM.

3. Read tilt

   Combine accelerometer angle and gyro rate with a complementary filter (98% gyro, 2% accel).

4. Run the PID loop

   Every 5 ms compute error = target angle − current angle and set motor speed from P, I and D terms.

5. Tune patiently

   Start with only P until it oscillates, add D to damp, then a little I to correct drift.

Code

# PID core (MicroPython) — read_angle() returns filtered tilt in degrees
import time
Kp, Ki, Kd = 25.0, 0.4, 0.8
target, integral, prev = 0.0, 0.0, 0.0
DT = 0.005

while True:
    angle = read_angle()
    err = target - angle
    integral = max(-300, min(300, integral + err * DT))
    deriv = (err - prev) / DT; prev = err
    out = Kp * err + Ki * integral + Kd * deriv
    set_motors(max(-1, min(1, out / 100)))
    if abs(angle) > 45: set_motors(0)        # fell over: stop
    time.sleep(DT)

Tip: test every script on the bench before installing it in a moving machine.

Questions