Related Experiment Video
Updated: Jul 13, 2026
04:15
Online Virtual Reality Networked Control Laboratory Applied in Control Engineering Education
Published on: February 23, 2024
Neural network control for position tracking of a two-axis inverted pendulum system: experimental studies
Seul Jung1, Hyun-Taek Cho, T C Hsia
1Intelligent Systems and Emotional Engineering Laboratory, Chungnam National University, Daejeon 305-764, Korea. jungs@cnu.ac.kr
IEEE Transactions on Neural Networks
|August 3, 2007
Summary
This study demonstrates a decentralized neural network control system for a 2-DOF inverted pendulum. The novel approach successfully achieved precise position tracking and stable angle balancing for the pendulum and cart system.
Area of Science:
- Robotics and Control Systems
- Artificial Intelligence
- Mechanical Engineering
Background:
- Traditional control methods struggle with complex, coupled dynamics in multi-DOF systems.
- Decentralized control offers potential for improved robustness and adaptability.
- Neural networks provide powerful tools for learning and adapting to system uncertainties.
Purpose of the Study:
- To present experimental validation of a decentralized neural network control scheme for a 2-DOF inverted pendulum.
- To investigate the effectiveness of reference compensation technique for decoupled control.
- To evaluate the system's ability to perform trajectory tracking while maintaining pendulum stability.
Main Methods:
- Implementation of a decentralized control structure with separate neural network controllers for each axis.
- Application of neural network controllers for both angle balancing and cart position tracking.
- Testing a circular trajectory tracking task to assess performance under coupled dynamics.
- Utilizing a reference compensation technique to manage system uncertainties and coupling effects.
Main Results:
- Successful experimental demonstration of the decentralized neural network control system.
- Achieved precise position tracking control for the cart on the x-y plane.
- Maintained stable pendulum angle control during trajectory tracking.
- The decoupled control structure effectively compensated for uncertainties and coupling.
Conclusions:
- The proposed decentralized neural network control scheme is effective for controlling a 2-DOF inverted pendulum.
- The reference compensation technique enhances robustness and performance in complex robotic systems.
- Experimental results confirm the system's capability for accurate trajectory tracking and stable balancing.
Related Concept Videos
01:09Torsional Pendulum
A torsional pendulum involves the oscillation of a rigid body in which the restoring force is provided by the torsion in the string from which the rigid body is suspended. Ideally, the string should be massless; practically, its mass is much smaller than the rigid body's mass and is neglected.
As long as the rigid body's angular displacement is small, its oscillation can be modeled as a linear angular oscillation. The amplitude of the oscillation is an angle. The role of mass is played by the...
As long as the rigid body's angular displacement is small, its oscillation can be modeled as a linear angular oscillation. The amplitude of the oscillation is an angle. The role of mass is played by the...
01:21Linear Approximation in Time Domain
Nonlinear systems often require sophisticated approaches for accurate modeling and analysis, with state-space representation being particularly effective. This method is especially useful for systems where variables and parameters vary with time or operating conditions, such as in a simple pendulum or a translational mechanical system with nonlinear springs.
For a simple pendulum with a mass evenly distributed along its length and the center of mass located at half the pendulum's length, the...
For a simple pendulum with a mass evenly distributed along its length and the center of mass located at half the pendulum's length, the...
01:10Simple Pendulum
A simple pendulum consists of a small diameter ball suspended from a string, which has negligible mass but is strong enough to not stretch. In our daily life, pendulums have many uses, such as in clocks, on a swing set, and on a sinker on a fishing line.
The period of a simple pendulum depends on two factors: its length and the acceleration due to gravity. The period is completely independent of any other factors, such as mass or maximum displacement. For small displacements, a pendulum is...
The period of a simple pendulum depends on two factors: its length and the acceleration due to gravity. The period is completely independent of any other factors, such as mass or maximum displacement. For small displacements, a pendulum is...
01:06Physical Pendulum
When a rigid body is hanging freely from a fixed pivot point and is displaced, it oscillates similar to a simple pendulum and is known as a physical pendulum. The period and angular frequency of a physical pendulum are obtained by using the small-angle approximation and drawing parallels with a spring-mass system. The small-angle approximation (sinθ=θ) is valid up to about 14°.
When dealing with complicated systems, the mass moment of inertia is an important parameter, as it describes the mass...
When dealing with complicated systems, the mass moment of inertia is an important parameter, as it describes the mass...
01:27Real-World Applications of Power Series
The motion of a simple pendulum is governed by Newton’s Second Law in its rotational form, which relates the net torque on the bob to its angular acceleration. This physical law gives rise to a second-order differential equation in which the angular acceleration is proportional to the sine of the displacement angle.Because of the sin(𝜃) term, the governing equation is a nonlinear differential equation, which is difficult to solve analytically. To simplify the mathematical model, the sine...
01:24One-Degree-of-Freedom System
In mechanical engineering, one-degree-of-freedom systems form the basis of a wide range of electrical and mechanical components. Using these models, engineers can predict the behavior of various parts in a larger system, which gives them insight into how different forces interact with each other.
A one-degree-of-freedom system is defined by an independent variable that determines its state and behavior. One example of a one-degree-of-freedom system is a simple harmonic oscillator, such as a...
A one-degree-of-freedom system is defined by an independent variable that determines its state and behavior. One example of a one-degree-of-freedom system is a simple harmonic oscillator, such as a...