Source code for stonesoup.models.control.linear

# -*- coding: utf-8 -*-

import numpy as np
from scipy.stats import multivariate_normal

from .base import ControlModel
from ..base import LinearModel
from ...base import Property


[docs]class LinearControlModel(ControlModel, LinearModel): r"""Implements a linear effect to the state vector via, .. math:: \hat{x}_k = B_k \mathbf{u}_k + \gamma_k where :math:`B_k` is the control-input model matrix (i.e. control matrix), :math:`\mathbf{u}_k` is the control vector and :math:`\gamma_k` is sampled from zero-mean white noise distribution :math:`\mathcal{N}(0,\Gamma_k)` """ control_vector: np.ndarray = Property(doc="Control vector at time :math:`k`") control_matrix: np.ndarray = Property( doc="Control input model matrix at time :math:`k`, :math:`B_k`") control_noise: np.ndarray = Property( default=None, doc="Control input noise covariance at time :math:`k`") @property def ndim(self): return self.ndim_ctrl @property def ndim_ctrl(self): return self.control_vector.shape[0]
[docs] def matrix(self): """ Returns ------- : :class:`numpy.ndarray` the control-input model matrix, :math:`B_k` """ return self.control_matrix
[docs] def control_input(self): r"""The mean control input Returns ------- : :class:`numpy.ndarray` the noiseless effect of the control input, :math:`B_k \mathbf{u}_k` """ return self.control_matrix @ self.control_vector
[docs] def rvs(self): r"""Sample (once) from the multivariate normal distribution determined from the mean and covariance control parameters Returns ------- : :class:`numpy.ndarray` a sample from :math:`\mathcal{N}(B_k \mathbf{u}_k, \Gamma_k)` """ return multivariate_normal.rvs(self.control_input(), self.control_noise).reshape(-1, 1)
[docs] def pdf(self, control_vec): """The value of the probability density function (pdf) at a test point Parameters ---------- control_vec : :class:`numpy.ndarray` The control vector at the test point Returns ------- float The value of the pdf at :obj:`control_vec` """ return multivariate_normal.pdf(control_vec, mean=self.control_input(), cov=self.control_noise).reshape(-1, 1)