Accelerating the pace of engineering and science

Developing Multivariable Control Systems for Robotics

Shashank Prasanna, MathWorks

In this webinar, you will learn how to use simulation to design and implement multivariable controllers for a four-joint robot arm. We use two different techniques that go beyond the traditional tuning of individual PID controllers. The first method shows automatic tuning of four coupled PID controllers simultaneously. The second approach demonstrates the design of a model predictive controller (MPC) that uses an internal model of the robot arm to forecast future plant behavior and adjust control actions accordingly.

After we develop the two controllers, we test and verify their performance. The controllers first are verified against the nonlinear robot arm model using desktop simulation. Then, automatic code generation is used to implement the controllers in C code. Finally, we complete the design validation by controlling a robot arm in real time.

About the Presenter: Shashank Prasanna is an application support engineer at MathWorks. He has extensive experience with MATLAB and Simulink. Shashank holds an M.S. in electrical engineering with a major in control systems from Arizona State University.

Product Focus

  • Control System Toolbox
  • Embedded Coder
  • Model Predictive Control Toolbox
  • Robust Control Toolbox
  • Simulink Control Design
  • Simulink Real-Time

Recorded: 14 Feb 2013