Related Experiment Videos
Design and analysis of controllers for a double inverted pendulum
Henrik Niemann1, Jesper Kildegaard Poulsen
1Orsted-DTU, Automation, Technical University of Denmark, DK-2800 Lyngby, Denmark. hhn@oersted.dtu.dk
ISA Transactions
|February 3, 2005
Summary
This study explores physical control challenges using H-infinity and micro-control methodologies. It demonstrates how system limitations impact controller design for a double inverted pendulum on a cart.
Area of Science:
- Control Systems Engineering
- Robotics
- Applied Mathematics
Background:
- Physical control problems require robust methodologies to address system complexities.
- Modeling and uncertainty are critical aspects of controller design.
- The double inverted pendulum on a cart is a standard benchmark for control system evaluation.
Purpose of the Study:
- To investigate the application of H-infinity and micro-control methodologies to a physical control problem.
- To analyze the impact of system limitations on controller design.
- To discuss modeling, uncertainty, performance specification, and implementation aspects.
Main Methods:
- Utilized H-infinity control theory for robust controller design.
- Applied micro-control methodologies for system analysis.
- Conducted a laboratory experiment using a double inverted pendulum on a cart.
- Investigated system limitations including control signal constraints and physical movement restrictions.
Main Results:
- Demonstrated the influence of control signal limitations on controller performance.
- Showcased the effect of cart movement restrictions on system stability and design.
- Provided insights into the practical implementation challenges of advanced control techniques.
- Validated the theoretical controller designs through a physical experiment.
Conclusions:
- System performance limitations significantly affect the design choices for H-infinity and micro-controllers.
- Robust control design must explicitly account for physical constraints.
- The double inverted pendulum experiment provides a valuable platform for studying real-world control challenges.