System design: Agile programming streamlines software development
Agile programming helps identify and prioritize programming tasks in need of attention and address them one at a time. This involves some research, coding, and especially testing, and aims to streamline the code development lifecycle to enhance future machine designs.
In the lifecycle of a machine from inception through implementation, code can be effectively developed and tested through simulation. The right programming tools are essential to simplify design, especially as functionalities are integrated. Virtual reality machine design and modeling environment described below may be available in the future.
Future of machine design
After arriving at work, you position your chair and put on your virtual reality glasses and gloves. You start designing by asking for material such as “A10,” then stretching it to the desired size, continuing with other virtual parts until the whole machine is designed. The machine looks good as it rotates in front of you, but now comes the fun part. By selecting machine pieces and describing how they move, you can simulate how the machine will operate on the plant floor. Once satisfied, a verbal command will yield assembly/detail drawings and code for the machine controller. The end product is also a model, not just a machine design. It can be used to predict operations when using different product elasticity or weight, for instance.
When virtual reality machine design and modeling is possible, you might feel like you’re in an aerobics class rather than an engineering office, but for now, we still use software products that are a little less sci-fi to accomplish our design goals. Development, simulation, and modeling software tools are generally implemented in separate packages. The package count increases because the engineering disciplines, such as mechanical, electrical, and control, frequently have their own dedicated software.
Goal: Easier programming
As a programmer, my primary goal is to make innovative machines that offer outstanding performance, and for this I use every trick in the book. One trick is to start with a framework that already has time-saving features such as error handling and recipe management. This frees up my time to concentrate on the machine-specific issues, and for those I use agile programming. The term “agile programming” means identifying and prioritizing programming tasks that are in need of attention and addressing them one at a time. This involves some research, coding, and especially testing.
Users can simplify research, coding, and testing with Beckhoff’s TwinCAT software, to develop automation system code and test it by simulating how it will work on production machinery. This results in proven code that’s ready to automate the machine as soon as it’s assembled.
After all, machine start-up should be a commissioning effort by adjusting sensors and timers, rather than a software debug effort with major code rewrites. The software also can assist in “rapid control prototyping,” which consists of several steps:
- Creation of a model of the system to be automated
- Design of the controller
- Simulation of the entire system
- Implementation of the controller design on the target platform
- Test of the implementation by coupling the target platform to the system model, “software in the loop” (SIL)
- Test of the real system with the real controller.
Tools can assist in simulation. The TwinCAT Simulation Manager simplifies configuration of a simulation environment, which integrates into the TwinCAT system environment. This tool supports the creation of a “virtual machine,” which corresponds to a real one in its runtime performance. In the lifecycle of a machine from inception through implementation, the right programming tools are essential.
- Gary Kirckof is application engineer manager at Beckhoff Automation. Edited by Mark T. Hoske, content manager, CFE Media, Control Engineering, mhoske(at)cfemedia.com.