用C++实现的PID控制算法（摘自一本书）

monkhooder

#include <cmath>

// Select 'double' or 'float' here:
typedef double real;
class PID_Controller
{
public:
    void Initialize(real kp, real ki, real kd,
        real error_thresh, real step_time);
    real Update(real error);

private:
    bool m_started;
    real m_kp, m_ki, m_kd, m_h, m_inv_h, m_prev_error,
        m_error_thresh, m_integral;
};

void PID_Controller::Initialize(real kp, real ki,
    real kd, real error_thresh, real step_time)
{
    // Initialize controller parameters
    m_kp = kp;
    m_ki = ki;
    m_kd = kd;
    m_error_thresh = error_thresh;

    // Controller step time and its inverse
    m_h = step_time;
    m_inv_h = 1 / step_time;

    // Initialize integral and derivative calculations
    m_integral = 0;
    m_started = false;
}

real PID_Controller::Update(real error)
{
    // Set q to 1 if the error magnitude is below
    // the threshold and 0 otherwise
    real q;
    if (fabs(error) < m_error_thresh)
        q = 1;
    else
        q = 0;

    // Update the error integral
    m_integral += m_h*q*error;

    // Compute the error derivative
    real deriv;
    if (!m_started)
    {
        m_started = true;
        deriv = 0;
    }
    else
        deriv = (error - m_prev_error) * m_inv_h;

    m_prev_error = error;

    // Return the PID controller actuator command
    return m_kp*(error + m_ki*m_integral + m_kd*deriv);
}

monkhooder

使用时需要先选择是float类型还是double类型