Skip to content

Scaling difference in impulse response size compared to scipy.signal #813

Description

@sawyerbfuller

Discussed in #747

Originally posted by fhchl June 29, 2022
I had some code that I wanted to port from using scipy.signal.StateSpace etc. to python-control. There seems to be some scaling difference between the two implementations. What am I overlooking?

import numpy as np
import control
import scipy.signal

# random state space matrices
t= np.linspace(0, 10, dtype=float)
ss = control.drss(3, 2, 1)
A, B, C, D = ss.A, ss.B, ss.C, ss.D
dt = t[1]

# compute impulse response with python-control
ss = control.ss(A, B, C, D, dt=dt)
t_control, h_control = control.impulse_response(ss, T=t)
plt.figure()
plt.plot(t_control, h_control[0, 0])
plt.plot(t_control, h_control[1, 0])

# compute impulse response with scipy.signal
ss_scipy = scipy.signal.StateSpace(A, B, C, D, dt=dt)
t_scipy, (h_scipy,) = scipy.signal.dimpulse(ss_scipy, t=t)
plt.figure()
plt.plot(t_scipy, h_scipy[:, 0])
plt.plot(t_scipy, h_scipy[:, 1])

plt.show()

Creates these two here.

output

output

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions