Chapter 8: Control Systems for Humanoids

Overview

What You'll Learn: Control theory basics, PID controllers, ROS 2 control framework, trajectory generation

Estimated Time: 7-9 hours

Core Concepts

PID Control

class PIDController:
    def __init__(self, kp, ki, kd):
        self.kp = kp  # Proportional gain
        self.ki = ki  # Integral gain
        self.kd = kd  # Derivative gain

        self.prev_error = 0.0
        self.integral = 0.0

    def update(self, setpoint, measurement, dt):
        error = setpoint - measurement

        # P term
        p_term = self.kp * error

        # I term
        self.integral += error * dt
        i_term = self.ki * self.integral

        # D term
        derivative = (error - self.prev_error) / dt
        d_term = self.kd * derivative

        # Update
        self.prev_error = error

        # Control output
        output = p_term + i_term + d_term
        return output

ROS 2 Control Architecture

Trajectory Generation

from trajectory_msgs.msg import JointTrajectory, JointTrajectoryPoint

def generate_trajectory(joint_names, positions, duration):
    traj = JointTrajectory()
    traj.joint_names = joint_names

    point = JointTrajectoryPoint()
    point.positions = positions
    point.time_from_start = Duration(seconds=duration)

    traj.points.append(point)
    return traj

Lab: PID Tuning

Objective: Implement and tune PID controller for joint position control

Tasks:

1. Implement PID class in Python
2. Create test node that commands joint positions
3. Tune Kp, Ki, Kd gains for stable control
4. Plot response (setpoint vs. actual)

Summary

Key Takeaways:

1. PID provides feedback control for position/velocity tracking
2. ROS 2 Control standardizes controller interfaces
3. Trajectory generation creates smooth motion plans
4. Controller tuning critical for stability and performance

Next: Part III: Gazebo & Unity