Customer Solutions

PosiCon Ball – A Laboratory Experiment to Teach Control Systems Design

Author(s):

Wolfgang Werth, University of Applied Sciences School of Electronics

Industry:

University/Education

Product:

Data Acquisition, LabVIEW

The Challenge:

Devising a teaching method to introduce students to control systems analysis and design, taking into account general physics, circuit analysis, and mathematics.

The Solution:

Using NI software and hardware products to develop an easy to understand laboratory experiment called PosiCon Ball, which introduced students to the challenges of real-world control systems through signal measurement and control concept implementation.

Laboratory Experiment PosiCon Ball

Motivating Students to Learn Control Systems Design
Clear laboratory experiments are some of the most powerful ways to teach students complex technical materials. Because understanding control systems is a difficult and time-consuming task to many students, we (with a student group) constructed a flexible, powerful, cost-effective, real-world control experiment called PosiCon Ball. We used NI LabVIEW to implement and test different controllers and NI hardware components to manage the data acquisition and control tasks.

Control engineering involves designing and building real physical systems to perform given tasks. Control systems analysis and design is sometimes a very difficult, time-consuming task to many students because it requires a background knowledge of general physics, circuit analysis, mathematics, programming, and data acquisition. Within the last few years, many interesting but expensive laboratory experiments have become available. However, they are generally limited in flexibility and power, and are constrained to a given control platform.

To counteract these challenges, we developed a graphic, easily extensible, powerful control experiment called PosiCon Ball. Its main task is to control position x of a ball balancing on a moveable beam. PosiCon Ball consists of all of the important components of a typical control system – plant (mechanical construction including the movement of the ball), power unit, actuator (DC motor), sensors (potentiometer foil) and control unit (PC, microcontroller). We believe that, even if the plant represents a nonlinear, unstable dynamical system, its control task (the idea of how a good controller should react) should be easy to understand, even for undergraduate students.

How PosiCon Ball Works
With PosiCon Ball, a ball travels along a moveable beam. If the beam is positioned horizontally, the ball does not accelerate. But by manipulating the beam angle (a), we can manipulate the ball position. This is demonstrated in the following equation:

x = –5/7 g sin(a)

This movement is generated when we make the appropriate angle adjustment using a DC motor. An appropriate actuator transforms the rotary motor movement into a vertical beam movement (y). A wear-resistant potentiometer foil measures the ball position, and an incremental counter detects the motor angle. The data acquisition board easily captures the corresponding signals (analog voltage, digital counter signal). A pulse-width modulation unit driven by the DAQ-board moves the DC motor.

The system includes a second ball and beam unit which generates a specific reference signal when the user manually moves the reference ball. The control task is thereby done by a person, who moves the beam manually to perform the desired ball placement and acts as a controller.

System Design and Performance
To automatically perform the positioning task, students must design an appropriate controller using appropriate mathematical modeling. In a guided prelaboratory exercise. students test different (digital) control concepts such as:

• State-space concept
• State-space concept with integrator
• Cascade concept (lead-compensator, PID controller)

The control law is represented by a differential equation. For example, in the case of a state-space controller, we have:

uk = kT xk + Vrk

The system uses LabVIEW as a development tool to simplify controller implementation. Students can use the front panel to perform several different control methods. They can flexibly implement the control law by a formula node, meaning that they can change specific parameters or even the control law structure using LabVIEW case structure. In addition to the LabVIEW implementation, students can perform the control task with the C167 microcontroller unit programmed in C. Using this approach, students can use the C-code in the LabVIEW formula node without making many changes.

The system controller performs many tasks besides calculating the desired actuating signal. It controls the selected controller mode (PC or mC). Slider movement in the panel (instead of the reference signal unit) controls the ball position. To guarantee power, we implemented safety arrangements including motor current and voltage limitation. Furthermore, in designing the formula node, we took into account how the controller would react if the ball falls down from the beam. Besides the tracking problem, the model easily demonstrates disturbance rejection.

Designing for the Future
Using NI products, we built a flexible, user-friendly, cost-effective control system. NI products provide a perfect solution for our control experiment because of the control platform flexibility. For example, we could use an NI PXI-based system to control behavior under real-time conditions without any changes at the control unit.

Students continue to perform a wide variety of new experiments to improve their control engineering knowledge. For example, a current student project uses NI vision products for image detection to control ball position.

For more information, contact:
Wolfgang Werth
Carinthia Tech Institute
Fachhochschule Technikum Kärnten
Technologiepark Villach
Europastrasse 4
A-9524 Villach/St. Magdalen
Tel.: +43 (0)4242 90500-0
Fax: +43 (0)4242 90500-2110