North American X‑15 — продольная динамика¶

Экспериментальный ракетоплан X‑15 для исследований гиперзвукового полёта. Страница оформлена по аналогии с ELV: быстрый старт, математика, производные и API.

Быстрый старт

Запустите среду или модель за минуты.

К примеру
API модели

Документация Python‑класса X‑15.

К API
Среда Gymnasium

Готовая среда для RL‑агентов.

К среде
Теория

Уравнения состояния и численные параметры.

К модели

Как устроен объект управления¶

Модель задана в пространстве состояний:

\[\dot{x} = A x + B u, \quad y = C x + D u\]

Где:

\[ x = \begin{bmatrix} u & \alpha & q & \theta \end{bmatrix}^{\top}, \quad u_{in} = \eta \]

О единицах измерения

Углы и угловые скорости — в радианах. Методы API поддерживают выдачу в градусах.

Математическая модель¶

\[ \dot{x} = A x + B u, \qquad y = C x + D u \]

Численные матрицы (пример линеаризации):

\[ \begin{bmatrix} \dot{u} \\ \dot{\alpha} \\ \dot{q} \\ \dot{\theta} \end{bmatrix} = \begin{bmatrix} -0.0087 & -0.0190 & 0 & -32.174 \\ 0.0117 & -0.3110 & 1936 & 0 \\ 0.000471 & -0.0067 & -0.1820 & 0 \\ 0 & 0 & 1 & 0 \end{bmatrix} \begin{bmatrix} u \\ \alpha \\ q \\ \theta \end{bmatrix} + \begin{bmatrix} 0.0129 \\ -0.1844 \\ -0.0001225 \\ 0 \end{bmatrix} \eta \]

Производные (численные значения)¶

Матрица A (производные):

Коэффициент	Значение
x_u	-0.0087
x_α	-0.0190
x_q	0
x_θ	-32.174
z_u	0.0117
z_α	-0.3110
z_q	1936
z_θ	0
m_u	0.000471
m_α	-0.0067
m_q	-0.1820
m_θ	0

Вход η (столбец B):

Коэффициент	Значение
x_η	0.0129
z_η	-0.1844
m_η	-0.0001225

Источники¶

Heffley R. K., Jewell W. F. Aircraft handling qualities data. – NASA, 1972. № AD‑A277031.
Etkin B., Reid L. D. Dynamics of flight. – New York : Wiley, 1959. – Т. 2

Награда¶

Функция награды по умолчанию возвращает отрицательную абсолютную ошибку отслеживания угла тангажа:

\[r_t = -|\theta(t) - \theta_{\text{ref}}(t)|\]

Чем выше награда (ближе к 0), тем лучше качество отслеживания. Пользовательская функция награды может быть передана через параметр reward_func.

Быстрый старт¶

GymnasiumТолько модель

import gymnasium as gym 
import numpy as np

from tensoraerospace.envs import LinearLongitudinalX15
from tensoraerospace.utils import generate_time_period
from tensoraerospace.signals.standard import unit_step

dt = 0.01
tp = generate_time_period(tn=20, dt=dt)
number_time_steps = len(tp)
reference_signals = unit_step(degree=5, tp=tp, time_step=10, output_rad=True).reshape(1, -1)

env = gym.make(
    'LinearLongitudinalX15-v0',
    number_time_steps=number_time_steps, 
    initial_state=[[0],[0],[0],[0]],
    reference_signal=reference_signals,
)
state, info = env.reset()
for _ in range(200):
    action = np.array([[0.1]])
    state, reward, terminated, truncated, info = env.step(action)
    if terminated or truncated:
        break

import numpy as np
from tensoraerospace.aerospacemodel import LongitudinalX15

dt = 0.01
number_time_steps = 200

x0 = np.array([0.0, 0.0, 0.0, 0.0])

model = LongitudinalX15(
    x0=x0,
    number_time_steps=number_time_steps,
    selected_state_output=["u", "alpha", "q", "theta"],
    dt=dt,
)

for t in range(number_time_steps - 1):
    u = np.array([[0.05]])
    x_next = model.run_step(u)

Python API¶

МодельСреда Gymnasium

`LongitudinalX15(x0, number_time_steps, selected_state_output=None, t0=0, dt=0.01)` ¶

Bases: ModelBase

North American X-15 in longitudinal control channel.

.. warning:: This linearized model uses the FPS (foot-pound-second) unit system, NOT SI. The A/B matrices are taken directly from the reference Simulink source tensoraerospace/aerospacemodel/simulinkModel/x15/x15_data.m where g = 32.174 ft/s^2 and the trim velocity U0 = 1936 ft/s (~Mach 1.8 at altitude). Linear velocities u and w are perturbations in ft/s, and the gravity term A[0,3] = -32.174 is in ft/s^2.

If SI-valued states are fed in (e.g. ``u`` in m/s), the dynamics
will be silently inconsistent with the matrices. Callers that
operate in SI must convert (1 m/s = 3.28084 ft/s) at the
environment boundary.

Parameters:

Name	Type	Description	Default
`x0`	`ndarray \| list[float]`	Initial state of the control object.	required
`number_time_steps`	`int`	Number of time steps.	required
`selected_state_output`	`optional`	Selected states of the control object. Defaults to None.	`None`
`t0`	`int`	Initial time. Defaults to 0.	`0`
`dt`	`float`	Discretization frequency. Defaults to 0.01.	`0.01`

Action space

ele: elevator [rad] (input saturation at |ele| <= 25 deg, rate limit 60 deg/s, both converted to radians internally).

State space (FPS units): u: Longitudinal aircraft velocity perturbation [ft/s] w: Normal aircraft velocity perturbation [ft/s] q: Pitch angular velocity [rad/s] theta: Pitch angle [rad]

Output space (FPS units): u: Longitudinal aircraft velocity perturbation [ft/s] w: Normal aircraft velocity perturbation [ft/s] q: Pitch angular velocity [rad/s] theta: Pitch angle [rad]

Initialize LongitudinalX15 instance.

Parameters:

Name	Type	Description	Default
`x0`	`ndarray \| list[float]`	Initial state of the control object.	required
`number_time_steps`	`int`	Number of time steps.	required
`selected_state_output`	`list[str] \| None`	Selected states of the control object. Defaults to None.	`None`
`t0`	`float`	Initial time. Defaults to 0.	`0`
`dt`	`float`	Discretization frequency. Defaults to 0.01.	`0.01`

`import_linear_system()` ¶

Saved linearized matrices.

Units: matrices are in FPS (foot-pound-second) system, matching the reference Simulink file x15_data.m. In the A matrix the last column of the first row is -g = -32.174 ft/s^2 and A[1,2] = U0 = 1936 ft/s is the trim airspeed, not an SI (m/s) quantity. See the class docstring for a full unit list.

`initialise_system(x0, number_time_steps)` ¶

System initialization.

Parameters:

Name	Type	Description	Default
`x0`	`ndarray \| list[float]`	Initial state of the control object.	required
`number_time_steps`	`int`	Number of time steps in iteration.	required

`run_step(ut_0)` ¶

Execute one time step iteration.

Parameters:

Name	Type	Description	Default
`ut_0`	`ndarray`	Control vector.	required

Returns:

Type	Description
`ndarray`	np.ndarray: Control object state at step t+1.

`update_system_attributes()` ¶

Update attributes that change with each time step.

`get_state(state_name, to_deg=False, to_rad=False)` ¶

Get state array history.

Parameters:

Name	Type	Description	Default
`state_name`	`str`	State name.	required
`to_deg`	`bool`	Convert to degrees. Defaults to False.	`False`
`to_rad`	`bool`	Convert to radians. Defaults to False.	`False`

Returns:

Type	Description
`ndarray`	np.ndarray: Array of selected state history.

Example

state_hist = model.get_state('alpha', to_deg=True)

`get_control(control_name, to_deg=False, to_rad=False)` ¶

Get control signal array history.

Parameters:

Name	Type	Description	Default
`control_name`	`str`	Control signal name.	required
`to_deg`	`bool`	Convert to degrees. Defaults to False.	`False`
`to_rad`	`bool`	Convert to radians. Defaults to False.	`False`

Returns:

Type	Description
`ndarray`	np.ndarray: Array of selected control signal history.

Example

control_hist = model.get_control('stab', to_deg=True)

`get_output(state_name, to_deg=False, to_rad=False)` ¶

Get output signal array history.

Parameters:

Name	Type	Description	Default
`state_name`	`str`	Output signal name.	required
`to_deg`	`bool`	Convert to degrees. Defaults to False.	`False`
`to_rad`	`bool`	Convert to radians. Defaults to False.	`False`

Returns:

Type	Description
`ndarray`	np.ndarray: Array of selected output signal history.

Example

output_hist = model.get_output('alpha', to_deg=True)

`plot_output(output_name, time, lang='rus', to_deg=False, to_rad=False, figsize=(10, 10))` ¶

Plot output signal.

Parameters:

Name	Type	Description	Default
`output_name`	`str`	Output signal name.	required
`time`	`ndarray`	Time array.	required
`lang`	`str`	Label language ('rus' or 'eng'). Defaults to "rus".	`'rus'`
`to_deg`	`bool`	Convert to degrees. Defaults to False.	`False`
`to_rad`	`bool`	Convert to radians. Defaults to False.	`False`
`figsize`	`tuple`	Figure size. Defaults to (10, 10).	`(10, 10)`

Returns:

Type	Description
`Figure`	plt.Figure: Matplotlib figure object.

Raises:

Type	Description
`Exception`	If both to_rad and to_deg are specified, or if output_name is invalid.

Example

fig = model.plot_output('alpha', time_array, lang='rus', to_deg=True)

`LinearLongitudinalX15(initial_state, reference_signal, number_time_steps, tracking_states=None, state_space=None, control_space=None, output_space=None, reward_func=None)` ¶

Bases: Env

Simulation of LongitudinalX15 in OpenAI Gym for training agents.

.. warning:: The underlying :class:LongitudinalX15 uses FPS units (ft, lb, s). Linear velocities in the state vector are in ft/s, not m/s.

State vector: [u, w, q, theta] (FPS units) where: u - longitudinal velocity perturbation (ft/s) w - normal velocity perturbation (ft/s) q - pitch rate (rad/s) theta - pitch angle (rad)

Parameters:

Name	Type	Description	Default
`initial_state`	`ndarray \| list[float]`	Initial state.	required
`reference_signal`	`ndarray \| Callable`	Reference signal.	required
`number_time_steps`	`int`	Number of simulation steps.	required
`tracking_states`	`list[str] \| None`	Tracked states.	`None`
`state_space`	`list[str] \| None`	State space.	`None`
`control_space`	`list[str] \| None`	Control space.	`None`
`output_space`	`list[str] \| None`	Full output space (including noise).	`None`
`reward_func`	`Callable \| None`	Reward function (WIP status).	`None`

`reward(state, ref_signal, ts)` `staticmethod` ¶

Evaluate control performance.

Parameters:

Name	Type	Description	Default
`state`	`_type_`	Current state.	required
`ref_signal`	`_type_`	Reference state.	required
`ts`	`_type_`	Time step.	required

Returns:

Name	Type	Description
`reward`	`float`	Control performance evaluation.

`step(action)` ¶

Execute a simulation step.

Parameters:

Name	Type	Description	Default
`action`	`ndarray`	Array of control signals for selected control surfaces.	required

Returns:

Name	Type	Description
`next_state`	`ndarray`	Next state of the control object.
`reward`	`ndarray`	Evaluation of control algorithm actions.
`done`	`bool`	Simulation status, whether completed or not.
`logging`	`any`	Additional information (not used).

`reset(seed=None, options=None)` ¶

Reset simulation environment to initial conditions.

Parameters:

Name	Type	Description	Default
`seed`	`int`	Seed for random number generator.	`None`
`options`	`dict`	Additional options for initialization.	`None`

`render()` ¶

Visual display of actions in the environment. Status: WIP.

Raises:

Type	Description
`NotImplementedError`	Rendering is not available.

North American X‑15 — продольная динамика¶

Как устроен объект управления¶

Математическая модель¶

Производные (численные значения)¶

Источники¶

Награда¶

Быстрый старт¶

Python API¶

LongitudinalX15(x0, number_time_steps, selected_state_output=None, t0=0, dt=0.01) ¶

import_linear_system() ¶

initialise_system(x0, number_time_steps) ¶

run_step(ut_0) ¶

update_system_attributes() ¶

get_state(state_name, to_deg=False, to_rad=False) ¶

get_control(control_name, to_deg=False, to_rad=False) ¶

get_output(state_name, to_deg=False, to_rad=False) ¶

plot_output(output_name, time, lang='rus', to_deg=False, to_rad=False, figsize=(10, 10)) ¶

LinearLongitudinalX15(initial_state, reference_signal, number_time_steps, tracking_states=None, state_space=None, control_space=None, output_space=None, reward_func=None) ¶

reward(state, ref_signal, ts) staticmethod ¶

step(action) ¶

reset(seed=None, options=None) ¶

render() ¶

`LongitudinalX15(x0, number_time_steps, selected_state_output=None, t0=0, dt=0.01)` ¶

`import_linear_system()` ¶

`initialise_system(x0, number_time_steps)` ¶

`run_step(ut_0)` ¶

`update_system_attributes()` ¶

`get_state(state_name, to_deg=False, to_rad=False)` ¶

`get_control(control_name, to_deg=False, to_rad=False)` ¶

`get_output(state_name, to_deg=False, to_rad=False)` ¶

`plot_output(output_name, time, lang='rus', to_deg=False, to_rad=False, figsize=(10, 10))` ¶

`LinearLongitudinalX15(initial_state, reference_signal, number_time_steps, tracking_states=None, state_space=None, control_space=None, output_space=None, reward_func=None)` ¶

`reward(state, ref_signal, ts)` `staticmethod` ¶

`step(action)` ¶

`reset(seed=None, options=None)` ¶

`render()` ¶