Scribd Logo
Upload

Welcome to Scribd!

  • 47 views

    Swing-Up Control of Inverted Pendulum: B.Tech Pre-Project Presentation

    Uploaded by

    Daniyal Wahid
    The document summarizes a student project to design and implement a swing-up controller for an inverted pendulum mounted on a cart. The objectives are to design nonlinear and linear controllers to swing the pendulum up to the unstable vertical position and stabilize it there. The system will be modeled using Euler-Lagrange equations and linearized about the equilibrium points. Control approaches like pole placement and LQR will be considered. A timeline outlines completing literature review, simulation, and building the physical system over the next 6 months.

    Copyright:

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

    Swing-Up Control of Inverted Pendulum: B.Tech Pre-Project Presentation

    Uploaded by

    Daniyal Wahid
    47 views16 pages
    The document summarizes a student project to design and implement a swing-up controller for an inverted pendulum mounted on a cart. The objectives are to design nonlinear and linear controllers to swing the pendulum up to the unstable vertical position and stabilize it there. The system will be modeled using Euler-Lagrange equations and linearized about the equilibrium points. Control approaches like pole placement and LQR will be considered. A timeline outlines completing literature review, simulation, and building the physical system over the next 6 months.

    Original Description:

    Swing up control of a cart inverted Pendulum setup

    Original Title

    Inverted_Pendulum

    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 summarizes a student project to design and implement a swing-up controller for an inverted pendulum mounted on a cart. The objectives are to design nonlinear and linear controllers to swing the pendulum up to the unstable vertical position and stabilize it there. The system will be modeled using Euler-Lagrange equations and linearized about the equilibrium points. Control approaches like pole placement and LQR will be considered. A timeline outlines completing literature review, simulation, and building the physical system over the next 6 months.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    47 views16 pages

    Swing-Up Control of Inverted Pendulum: B.Tech Pre-Project Presentation

    Uploaded by

    Daniyal Wahid
    The document summarizes a student project to design and implement a swing-up controller for an inverted pendulum mounted on a cart. The objectives are to design nonlinear and linear controllers to swing the pendulum up to the unstable vertical position and stabilize it there. The system will be modeled using Euler-Lagrange equations and linearized about the equilibrium points. Control approaches like pole placement and LQR will be considered. A timeline outlines completing literature review, simulation, and building the physical system over the next 6 months.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 16

    SWING-UP CONTROL OF INVERTED

    PENDULUM
    B.Tech pre-project presentation

    Yawer Nabi, Daniyal Wahid, Sheikh Kifayat


    Supervised by
    Dr. S.A. Nahvi
    Department of Electrical Engineering
    Islamic University of Science and Technology

    May 17, 2021

    1/16
    PRESENTATION OUTLINE

    1. Introduction
    System Description

    2. Objectives
    3. Literature review
    4. System modelling
    5. Tentative timeline
    6. References

    2/16
    Introduction
    • Control design of a cart-inverted pendulum system is a benchmark problem in
    control systems.
    • The system consists of an inverted pendulum on a motorized cart. Both the
    pendulum and the cart have one degree of freedom each.

    Figure: Typical representation of the system


    Introduction 3/16
    System Description

    • The pendulum has two equilibrium points, θ = 0 is stable and θ = π is unstable.


    • The system dynamics are non-linear.
    • The control input is force on the cart.
    • The outputs are angular position of the pendulum and horizontal position of the
    cart.
    • As the cart moves, the pendulum oscillates between 0 and π.

    Introduction 4/16
    Objectives

    • We propose to design and fabricate a swing-up controller for an inverted


    cart-pendulum system.
    • This would require designing two controllers, a nonlinear controller to swing-up the
    pendulum and a linearising controller to stabilize it at the position of unstable
    equilibrium.

    Objectives 5/16
    Literature review

    1.Since the 1950’s, the cart inverted pendulum was used for teaching linear feedback
    control theory[4] to stabilize open-loop unstable systems.
    2.The inverted pendulum benchmark has been considered in many references as a typical
    root-locus analysis example [4].
    3. It has been also used in many books to solve the Linear optimal control problem[4].
    4. Also used to solve complex non-linear control problem[4] for unstable systems.
    5. In recent years, the need of analysis and control of under-actuated robotic systems
    arose in many practical applications. The cart inverted pendulum benchmark is a typical
    example of such systems[1].
    6. This problem has also been integrated to stabilise a vehicle roll-over[2]

    Literature review 6/16


    System modelling

    System modelling is done using Euler-Lagrangian equations.


    d ∂L ∂L
    ( )− = Qi (1)
    dt ∂qi ∂qi

    Symbols used
    u = input force, x = cart displacement, θ = pendulum angle from the vertical, KE =
    kinetic energy of the system, PE = potential energy of the system, L = KE − PE

    System modelling 7/16


    Continued

    EOM can be derived by substituting Lagrangian into Euler-Lagrange equation.


    The total kinetic energy of system is expressed by
    1 1
    KE = (M + m)ẋ 2 + (ml 2 θ̇ − 2ml ẋ θ̇cosθ) (2)
    2 2
    The total potential energy of the system is

    PE = mglcosθ. (3)

    System modelling 8/16


    Non Linear 2nd Order Differential equations
    By solving, the equations of motion come out to be
    (M + m)ẍ − ml θ̇cosθ + ml θ̇sinθ = F (4)
    ml θ̈ − mgsinx − mẍcosθ = 0 (5)
    This model should be linearized.
    Linearization is performed about equilibrium points.
    θ=π
    θ=0
    The linearized equations are:
    (M + m)ẍ + ml θ̈ + b ẋ = F (6)
    I θ̇ + ẍ = g θ (7)
    System modelling 9/16
    State Space Representation

    The Linearised EOM can be represented in state-space form if re-arranged into a series
    of 1st ODE’s.
    The standard Matrix form is = Ax+Bu.
    x=state vector ; u=input vector ; A=state matrix ; B=input matrix and u=F.
          
    ẋ 0 1 0 0 x 0
    ẍ  0 −b −mg 0 ẋ   1 
     = M M   +  M u (8)
    θ̇  0 0 0 0  θ   0 
    b M+m −1
    θ̈ 0 Ml Ml
    g 0 θ̇ M

    System modelling 10/16


    Energy Based Swing Up Control

    1. For swing up control a non-linear controller has to be designed , because of the large
    displacement.
    2. This has application in preventing vehicle roll-over [2]
    3. Two controllers are to be designed.
    4. Non-linear to swing up to unstable position.
    5. Linear to stabilize at unstable equilibrium position.
    6. It has been demonstrated in literature [1,2], that swing up of cart pole can be
    achieved by partial feedback linearization and energy shaping.
    7. The problem is defined as applying an input force u to the cart that drives pendulum
    to a neighborhood of inverted position (θ = 0)

    System modelling 11/16


    State Space Control

    Mainly of two types:


    1
    POLE PLACEMENT TECHNIQUE

    2
    LQR CONTROL

    POLE PLACEMENET TECHNIQUE


    1. We design a control law so as states attain steady state values from any arbitrary
    initial value.
    2.Control input of the system is (u = −kx).

    System modelling 12/16


    3.Substituting the eqn in steady state model the closed loop system becomes
    ẋ = (A − bk)x.
    4.k has to be designed in a way that eigen values of (A − bk) must have negative real
    part.
    5.In this we design state feedback gain matrix such that it places poles at desired
    location.
    6.Pair (A, B) must be controllable for arbitrary pole placement.
    7.In MATLAB we use command k = place(A, B, p). Then we compute k . p is the
    column matrix of desired poles.
    8.We select different Eigen values to check stability, and check system response for
    different values.
    9. As eigen values become negative ,system response is aggressive .
    10.We select eigen values in such a way that system is realizable and cost is reduced.

    System modelling 13/16


    Tentative Timeline

    * March ,April : Relevant literature reading , Arduino programming.


    ** May, June : Designing and simulation of Energy based swing-up control.
    *** July,August and September:Hardware assembly; Application of Energy based swing
    up control,report writing and paper writing.Application of state feedback control.

    System modelling 14/16


    References

    1. Energy based control of a class of underactuated mechanical system by M.W.Swong.


    2. Stabilizing a vehicle rollover: An analogy of cart pole stabilization by S.C.Peters ,
    J.E.Bobrow , and Karln Legemma.
    3. https://ctms.engin.umich.edu/CTMS/index.php?aux=Home.
    4.The inverted pendulum : A Fundamental Benchmark in Control Theory and Robotics
    Olfa Boubaker INSAT, Centre Urbain Nord BP.676-1080 Tunis Cedex , Tunisia.

    System modelling 15/16


    The End

    You might also like