Control

The MCSimPython.control package contains a set of DP and Maneuvering controllers.

Basic Controllers

The module contains simple PD and PID controllers.

PD Controller

class MCSimPython.control.basic.PD(kp: list, kd: list)

Proportional-Derivative controller.

get_tau(eta, eta_d, nu, nu_d)

Calculate control loads.

Parameters:

  • eta (array_like) – Vessel pose in surge, sway and yaw.

  • eta_d (array_like) – Desired vessel pose in surge, sway and yaw (NED-frame).

  • nu (array_like) – Vessel velocity in surge, sway, and yaw (body-frame).

  • nu_d (array_like) – Desired vessel velocity in surge, sway and yaw (body-frame).

Returns:

tau – Controller load in surge, sway, and yaw (body-frame).

Return type:

array_like

set_kd(kd: list)

Set the derivative gain coefficients.

set_kp(kp: list)

Set the proportional gain coefficients.

PID Controller

class MCSimPython.control.basic.PID(kp: list, kd: list, ki: list, dt: float = 0.01)

Proportional-Derivative control with integral action.

get_tau(eta, eta_d, nu, nu_d)

Calculate control loads.

Parameters:

  • eta (array_like) – Vessel pose in surge, sway and yaw.

  • eta_d (array_like) – Desired vessel pose in surge, sway and yaw (NED-frame).

  • nu (array_like) – Vessel velocity in surge, sway, and yaw (body-frame).

  • nu_d (array_like) – Desired vessel velocity in surge, sway and yaw (body-frame).

Returns:

tau – Controller load in surge, sway, and yaw (body-frame).

Return type:

array_like

PI Controller

class MCSimPython.control.basic.PI(kp: list, ki: list, dt: float = 0.01)

Proportional control with integral action.

get_tau(eta, eta_d)

Calculate control loads.

Parameters:

  • eta (array_like) – Vessel pose in surge, sway and yaw.

  • eta_d (array_like) – Desired vessel pose in surge, sway and yaw (NED-frame).

Returns:

tau – Controller load in surge, sway, and yaw (body-frame).

Return type:

array_like

Direct Bias Compensation Controller

class MCSimPython.control.basic.DirectBiasCompensationController(kp: list, kd: list)

Bias estimate provided from the observer as direct compensation in a nominal PD control law.

get_tau(eta, eta_d, nu, nu_d, b)
Parameters:

  • eta (array_like) – Vessel pose in surge, sway and yaw.

  • eta_d (array_like) – Desired vessel pose in surge, sway and yaw (NED-frame).

  • nu (array_like) – Vessel velocity in surge, sway, and yaw (body-frame).

  • nu_d (array_like) – Desired vessel velocity in surge, sway and yaw (body-frame).

  • b (array_like) – Estimated bias in surge, sway and yaw (body-frame)

Note

The bias term b is defined in body-frame.

Adaptive Controller

class MCSimPython.control.adaptiveFS.AdaptiveFSController(dt, M, D, N=15)

Adaptive controller using a truncated Fourier series-based internal disturbance model in a “Model Reference Adaptive Control” (MRAC). The control law is determined by using LgV-backstepping, based on calculations and procedures done in (Fossen, 2021) and (Brørby, 2022).

To be implemented:

  • Improved tuning.

get_regressor(t)

Extract a (2*N+1)-dimensional regressor defined as [1 cos(w1*t) sin(w1*t) cos(w2*t) … sin(wN*t)]. Used in get_tau() to calculate control forces.

Parameters:

t (float) – Time

Returns:

regressor – (2N + 1) dimensional, time-dependent vector as defined above

Return type:

array_like

get_tau(eta, eta_d, nu, eta_d_dot, eta_d_ddot, t, calculate_bias=False)

Calculate controller output based on the adaptive update law.

Parameters:

  • eta (array_like) – [surge, sway, yaw] expressed in NED

  • eta_d (array_like) – Desired position expressed in NED

  • nu (array_like) – [u, v, r] expressed in body

  • eta_d_dot (array_like) – Desired velocity expressed in NED

  • eta_d_ddot (array_like) – Desired acceleration expressed in NED

  • t (float) – Time

  • Calculate_bias (bool) – Function will also return estimated bias if True. Set to False as default.

  • Output –

  • ------------ –

    • tau (3DOF): Calculated control forces

    • b_hat (3DOF): Estimated bias in surge, sway, yaw. Only returned if “calculate_bias == True”

set_freqs(w_min, w_max, N)

Customize the number and magnitude of wave frequencies to be included in the internal disturbance model.

UNIT IN RAD/S OR HZ?

Parameters:

  • w_min (float) – Lower bound frequency

  • w_max (float) – Upper bound frequency

  • N (int) – Number of components

Backstepping Controllers

In the MCSimPython.control.backstepping module there are controllers based on backstepping theory. Currently, only a backstepping controller for maneuvering purposes has been implemented.

class MCSimPython.control.backstepping.BacksteppingController(M, D, K1, K2)

Backsteppipng controller for maneuvering.

The controller should be used together with a path parameterization guidance/reference model.

See also

MCSimPython.guidance.path_parm.WayPointPathParam

Way-points path parameterization.

control_law(z1, z2, alpha, alpha_dot, nu)

Complete control law.

error_pos(eta, eta_d)

Deviation from desired pose in body-frame.

error_vel(nu, alpha)

Deviation from virtual control law.

lyapunov_grad_s(z1, eta, eta_d_s)

Lyapunov function gradient.

u(eta, nu, eta_d, eta_d_s, eta_d_s2, mu, u_s, u_s_dot, ddt_u_s)

Control command.

unit_gradient_update_law(u_s, mu, eta_d_s, lyapunov_gradient)

Unit-tangent gradient update law for path parameter (s).

virtual_control_dot(z1, z2, alpha, eta, eta_d, eta_d_s, eta_d_s2, u_s, u_s_dot, nu, s_dot, ddt_u_s)

Time derivative of virtual control law (alpha_dot).

virtual_control_law(z1, eta, eta_d_s, u_s)

Virtual control law (alpha).