Beth, Excellent article. This approach has a ton of potential for transforming the way that automation/machinery is developed, designed and deployed. It will be interesting to see how this area develops with control vendors. There are some solutions in the marketplace now but this will require different planning and thinking in terms of the machinery development process to gain traction. Interesting.
Thanks Al. I could totally see how this design approach could have huge ramifications for easing the development burden around automation and controls machinery given the high content of software and the complex movements. From all accounts, it's a learning process, however, and a very different development approach than traditional engineering workflows. There is definitely a commitment required to get training for engineers, not just on the tools and software, but on the modeling work itself and how to best adapt it into design processes.
Festo makes an excellent point here, and there are many, many other embedded developed who have the same dilemma: It's often difficult to optimize controller designs by building and testing on physical hardware. This article is going to be a keeper for many embedded developers.
Beth, model based designing and prototyping are very important mile stones in any of the defence and avionic projects. Math lab and simulink are two major software's used for simulation purpose in engineering background. In most of the prototyping projects, spiral models are followed where blocks are building over the existing one in an incremental form.
@Mydesign: You are right--the process is a major milestone, particularly in those industries where embedded systems lie at the heart of system designs. Beyond MATLAB and Simulink, any other tools that you are familiar with that are giving developers a jump on model-based design processes?
Nice post!! It gave the knowledge if how the equipments are developed. Model based designed is an aproach to eveolve a platform for communicating through the entire design process in process of development cycle. The development is carried out through development of a plant, developing a device to control a plant, simulate the device for plant and the plant itself, and lastly intergrating all these.
I think that Al is seeing an emerging trend by industrial automation suppliers to support Model-Based Design. Besides FESTO, B&R Automation, Beckhoff Automation, and Bachmann Electronics all have targets that accept ANSI C-code generated from Simulink models. Siemens, too, offers a means to port C-code to a PC-based controller. This connectivity helps machine builders perform control system design using simulation and implement on their controllers.
MapleSim, from Maplesoft, is a model-based design tool built on a foundation of symbolic computation technology. It handles all of the complex mathematics involved in the development of engineering models, including multi-domain systems, plant modeling, and control design. It is the only comprehensive modeling system built within a natively symbolic framework. Therefore, not only does it save model development time from months to days, but also avoids some of the worst sources of error and computational inefficiencies generated by traditional, numeric-based modeling tools. Leading automotive and aerospace manufacturers, electronics system designers and high-end robotic design engineers are using MapleSim in their work. You can read more about it here: www.maplesim.com
Responding to Tony, I don't think there is any question but that modelling and simulation can be tremendous assets in developing better and more efficient automation control solutions. The key is that it will require a different way of thinking from the past, and probably experimenting with software solutions and sample projects for engineers to get comfortable. Getting past the questions of whether it will be worth it is important, especially since many companies already have their systems modelled already. Good stuff.
One of the big advantage of model-based design centers on breaking the "toss it over the wall" mentality that companies have had for years. Various departments, from marketing down to production had "walls" that blocked communications and thus caused problems. Design changes sometimes got misinterpreted due to lack of proper documentation and lack of regulat communications. By clearly defining system requirements at the start and giving everyone access to them, model-based design techniques help companies complete designs faster and with fewer problems. And because engineers can actually execute the models, they see how circuits, mechanics, and software work together before they build anything.
But, model-based design requires a large commitment of time, talent, and loot and a company that wants to make the jump needs an evangelist who can "sell" others on the virtues and benefits of this approach to designs. Not all engineers will want to jump on the bandwagon, though.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.