Scribd Logo
Upload

Welcome to Scribd!

  • 35 views

    Controls Field Exam

    Uploaded by

    VigneshRamakrishnan
    The document describes two control system design problems: 1) Designing a linear quadratic regulator (LQR) to control the altitude of a satellite orbiting Earth using thrusters. The satellite dynamics are modeled by a differential equation. 2) Developing a state estimator for a stochastic system with linear dynamics and a nonlinear measurement function, subject to process and measurement noise. Challenges for different noise scenarios are considered.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Controls Field Exam

    Uploaded by

    VigneshRamakrishnan
    35 views2 pages
    The document describes two control system design problems: 1) Designing a linear quadratic regulator (LQR) to control the altitude of a satellite orbiting Earth using thrusters. The satellite dynamics are modeled by a differential equation. 2) Developing a state estimator for a stochastic system with linear dynamics and a nonlinear measurement function, subject to process and measurement noise. Challenges for different noise scenarios are considered.

    Original Description:

    [

    Original Title

    C_2014

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document describes two control system design problems: 1) Designing a linear quadratic regulator (LQR) to control the altitude of a satellite orbiting Earth using thrusters. The satellite dynamics are modeled by a differential equation. 2) Developing a state estimator for a stochastic system with linear dynamics and a nonlinear measurement function, subject to process and measurement noise. Challenges for different noise scenarios are considered.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    35 views2 pages

    Controls Field Exam

    Uploaded by

    VigneshRamakrishnan
    The document describes two control system design problems: 1) Designing a linear quadratic regulator (LQR) to control the altitude of a satellite orbiting Earth using thrusters. The satellite dynamics are modeled by a differential equation. 2) Developing a state estimator for a stochastic system with linear dynamics and a nonlinear measurement function, subject to process and measurement noise. Challenges for different noise scenarios are considered.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 2

    Controls Field Exam

    Question 1
    Consider a satellite orbiting the earth. The satellites distance from earth is denoted by r(t). The satellite
    is put on orbit such that its nominal distance to earth is R, and we would like to design a controller to
    regulate r(t) around R using the thrusters on the satellite. The dynamics of the satellite can be described the
    following differential equation:

    r(t) = 2 r(t) + u(t),
    r(t)2
    where is the angular velocity of the satellite and is a constant that depends on the mass of the satellite.
    We chose the orbit radius R and the angular velocity such that

    2 R3 = .

    (Please note that this is a grossly simplified model of a satellite orbiting earth. We arrive at this model after
    making various assumptions and simplifications. The behavior of this simple model may not closely match
    more complex models of satellite dynamics. This question is intended only to test your control design skills.
    It can be answered without any knowledge of satellite dynamics.)
    Please answer the following questions:

    1. Design a regulator that regulates the altitude of the satellite. Use the infinite-horizon LQR control
    design methodology. Equally weigh the deviation of r(t) and u(t) from their nominal values, and do
    not (directly) include the deviation of r(t)
    from its nominal value in the cost function. Please compute
    an LQR controller that fulfills this specification. For simplicity, suppose = 1.

    2. Design a regulator (for the nonlinear system) such that the closed-loop dynamics behaves exactly like
    a linear time-invariant system with two poles at 2. For simplicity, suppose = = R = 1.

    1
    Question 2
    Consider a stochastic first-order system with linear dynamics and nonlinear measurement given by

    x = x + w
    y = f (x) + v

    where

    12 ex 1
    f (x) = x5 x
    7 e +1
    The process noise w and measurement noise v are independent, zero mean, Gaussian white noise processes
    with intensities W and R, respectively. The initial state of the system is known to be x(0) = 0. A plot of
    f (x) is below. For reference, the slope of f at x = 0 is approximately 0.252, and f (x) crosses the x axis
    at approximately x = 2.48.

    f (x)

    Describe how you would develop a state estimator for this problem, and the problems you might en-
    counter. You might consider some limiting cases:

    1. W  1, R  W

    2. W  1, R  W

    3. W  1, R  W

    4. W  1, R  W

    You might also like