nutopy.ocp

class OCP(f0fun, ffun)

Bases: object

Parameters

• f0fun (callable) – Function defining the Lagrange cost

• ffun (callable) – Function defining the dynamics

Returns

An Optimal Control Problem (Lagrange type) object

f0(t, x, u, *args, **kwargs)

Parameters

• t (float) – Time

• x (array) – State

• u (array) – Control

Returns

The value of the Lagrange cost

f(t, x, u, *args, **kwargs)

Parameters

• t (float) – Time

• x (array) – State

• u (array) – Control

Returns

The value of the dynamics

class Hamiltonian(hfun, tvars=None)

Bases: object

Parameters

hfun (callable) – Hamiltonian function (must be tensorized up to order two)

Returns

A Hamiltonian object

property tvars

property tvars_args

classmethod fromOCP(o, ufun, p0=- 1.0)

Parameters

• o (OCP) – Optimal control problem

• ufun (callable) – Dynamic feedback control

• p0 (float) – Value of cost multiplier (default is -1, for normal case)

Returns

Generates a Hamiltonian from OCP and dynamic feedback control

vec(t, x, p, *args, **kwargs)

Parameters

• t (float) – Time

• x (array) – State

• p (array) – Costate

Returns

Symplectic gradient of the Hamiltonian. Tensorized to order 1.

class Flow(h, options=None)

Bases: object

Parameters

• h (Hamiltonian) – Hamiltonian object defining the flow

• options (dict) – Dictionary of options for the IVP solver

Returns

The Hamiltonian flow. The optional options are passed to the IVP solver and can be set dynamically.

See also

nutopy.ivp.Options

property h

property options

exception Error

Bases: Exception

Exceptions of the module:

Unimplemented IVPError

exception Unimplemented

Bases: nutopy.ocp.Error

Raised when trying to use a (not yet) implemented feature

exception IVPError(status, message)

Bases: nutopy.ocp.Error

Parameters

• status (int) – Termination status from IVP

• message (string) – Explanation from IVP

Returns

An exception raised when IVP integration is not successful

previous

nutopy.nle.solve

next

nutopy.tools