Payoffs from merging design and test (continued)

A continuation of an interview with James Tung, The MathWorks fellow and chief strategist, which appeared in the Febuary 2011 Viewpoint column.

By Larry Maloney, Contributing Editor -- Test & Measurement World, 2/1/2011 12:00:00 AM

JAMES TUNG The MathWorks Fellow and Chief Strategist The MathWorks Natick, MA   Jim Tung has nearly 30 years of experience in the technical computing software field. A 22-year veteran of The MathWorks, he held the positions of VP of marketing and VP of business development before assuming his current role focusing on business and technology strategy and analysis. Tung previously held marketing and sales-management positions at Lotus Development and Keithley DAS, a pioneering manufacturer of PC-based data-acquisition systems. Tung earned a bachelor's degree from Harvard University. Contributing editor Larry Maloney conducted a phone interview with Jim Tung on the use of software-analysis and modeling tools for test applications. Read the first part of this interview. To read past "Viewpoint" columns, go to www.tmworld.com/viewpoint. More from the February 2011 issue.

Read the first part of this interview.

Q: What are chief benefits of Simulink's new xPC Target Turnkey testing solution?
A: This is a fully-assembled, real-time testing solution for rapid prototyping of control systems and hardware-in-the-loop simulations. Marrying xPC Target from MathWorks with real-time target machines and I/O modules, this tool simplifies the time-consuming and resource-intensive process of evaluating software platforms, hardware technologies, and project requirements.

A typical application for this solution would be robotics, where you might want to prototype the control algorithm that manages the motor for a robot arm. Here, xPC Target Turnkey lets you very quickly develop your control algorithms, generate code for those algorithms, and run them on the xPC Target Turnkey hardware, which interacts with the real robotic arm. You can also experiment with various control algorithms to assess the behavior of the robotics.

In another instance, you might have a model of a robot arm, along with the motors and drives associated with it. If you have a Simulink model of that system, you can generate C code from that model and run it on xPC Target Turnkey to simulate the robot arm's behavior and test an embedded controller.

Q: Can you give an example of how Simulink SystemTest helps engineers reduce the need for hard prototypes?
A: A good example might be parameter sweeping of a design, where I might want to run a whole series of simulations that systematically change the parameters of the design to arrive at the optimum configuration or performance. In a control system, you might want to simulate different variations of gain in your control algorithm as a key parameter. SystemTest makes it very easy to set up such simulations and compare the results.

Q: How has your large user base stimulated innovation at MathWorks?
A: It's true that our company has a large customer base-more than 1 million users in over 100 countries. But the advantage to us, in terms of innovation, comes not so much from the size of our customer base as it does from its diversity. We have engineers and scientists from a broad range of industries, as well as professionals in fields like finance. Users in R&D labs are creating the next generation of control systems, while engineers at manufacturers use our products to design reliable products and systems. So, we get a great variety of input from users on the tools and techniques that they need to perform their jobs better.

As it turns out, many of the products that we developed initially to address the needs of one sector end up benefiting the entire user community. For example, our users in the financial services community have led the way in advancing the state-of-the-art in optimization techniques, and we have integrated such methods into our Optimization Toolbox and Global Optimization Toolbox that also serve other engineers. So, there is a lot of cross-pollination and synergy across our user community.

Q: What are the biggest customer challenges that will push further innovation in your company's products?
A: Much of it relates to the quality, reliability, and time-to-market concerns we discussed earlier, as well as to the growing complexity of the products and technologies that customers are working on. There is so much more data to analyze, and customers need to get maximum insight as quickly as possible from that data. Users are tackling designs that involve multiple domains, greater functionality, and more embedded software.

More and more, you see the blending of technologies in design, such as electronics and mechanical systems, as well as the integration of multiple techniques like controls, signal processing, and image processing. Traditionally, such designs were done in silos, with electrical engineers focusing on certain aspects, while mechanical engineers and control engineers took on other subsystems. Nowadays, more engineering teams want to integrate those diverse elements much earlier in the design and analyze system behavior in virtual models well before they get to hardware.

Q: What regions are driving growth for MathWorks?
A: MathWorks has had a strong global presence almost from its beginning in 1984. Over half of company's sales comes from outside the US, but we are really seeing broad-based growth across the world's regions, and in traditional and emerging markets. Helping to extend our reach globally in recent years are the new offices that we've opened in such countries as India, China, and Japan, where we are dealing both with local companies and with multinational companies with operations there.

Q: Will MathWorks continue to focus on products that can be applied to a wide range of industries, as opposed to creating niche products?
A: Certainly, we prefer to develop products that can be used across many applications and many industries, which helps create that cross pollination, but we also develop toolboxes for specific applications or industries, such as finance, bioinformatics, or vehicle networking. Still, these niche tools are developed to serve as add-ons or "on ramps" to our major products, such as Matlab and Simulink.