Software Behind the Wheel: How Software is Driving the Auto Industry into the Future

By November 11, 2017Automotive, Connected Cars, PLM
Autonomous Driving (Popular Mechanics)

It’s Not Your Father’s Oldsmobile

Automakers have been keeping a steady pace of technology innovation and manufacturing excellence for over a century. Since the big industrial breakthrough of the highly efficient assembly line, auto manufacturers continue to design and manufacture cars that are successively better, safer and cleaner. For many decades, the industry has been at the center of the US economic development, and, to many, a social icon.

But more recently, over the past decade or so, the iconic and seemingly stable industry has been in turmoil. It has been undergoing massive changes caused by the cumulative effect of rapid technology innovation, disruptive business models, new competitors, and emerging supply chain ecosystems whose full impact is not fully comprehended yet.

One of the most profound changes the auto industry is grappling with is the emerging notion of connected and autonomous cars. Most industry visionaries and practitioners portray a bold vision of a future in which cars, occupants, and cloud-based information and control systems communicate and exchange information in an omnipresent Internet of Things cloud. Cars are becoming part of the Internet, or, in today’s parlance, they are yet additional “things” in the Internet of Things (IoT).

Connected to your mobile device, a digital infrastructure, and a wealth of streamed cloud-based content and services, your car isn’t a just car anymore. It is an information center; it’s your wallet; it’s your office on wheels.  Cars are no longer self-contained independent systems designed to insulate and shelter you from your surroundings. Quite the opposite. Cars in the future will be constituents of a broad network and a conduit of a continuous torrent of information and services delivered by a fast-growing ecosystem.

Connected Cars Stress Product Development Tools

More than any technology revolution underway today, the connected car, and, eventually, autonomous driverless cars, pose unprecedented challenges for automotive designers and engineers.

Of course, software has been used in cars for is not a new phenomenon and has been around for several decades. The first production car to incorporate embedded software was the 1977 General Motors Oldsmobile Toronado, which had an electronic control unit (ECU) to control spark timing.

Developing simple control software in the earlier days of software-controlled automotive systems wasn’t a very onerous task, and informal methods and basic software skills were quite adequate. And even as the appetite for more complex software-controlled systems increased, designers managed through the process using a rudimentary software engineering environment built around open-source configuration management and bug tracking tools, augmented by the unavoidable plethora of spreadsheets and lengthy informal email conversations.

Nowadays, software-driven functionality is expanding quickly from controlling and sometimes replacing mechanical subsystems to serving a key role in defining customer experience and create product differentiation, as an increasing number of functional and user experience features are implemented in software, enabling unique, finely-tuned features to satisfy different consumer segments and diverse tastes.

The volume, variations, and complexity of embedded software is straining the organic ability of OEMs and suppliers that struggle to reskill and ramp up critical software development capabilities.

The obstacle isn’t in writing the software code itself. As companies adopt agile development methods and use modeling tools that generate quality code at a click of a mouse, much of the coding has become automated. Rather, managing the seemingly infinite number of features and feature-combinations, and validating the design against an ever-growing volume of complex and interdependent requirements is a monumental development burden, stressing the traditional product development methods and tools that have not kept pace with the increased complexity.

And the gap in between pace of innovation and increase in product complexity, and the capability of available design and validation tools continues to widen.

On the Road to Autonomous Vehicles

As autonomous driving capabilities continue to progress towards level 3 of autonomous vehicles, the role of electronic control systems that host advanced software-driven functionality, especially for advanced safety systems (ADAS) and autonomously-driven cars is expanding quickly. The need to incorporate additional sensors and increase the density of electronic controllers while struggling with physical space, weight and costs is driving innovation and experimentation with new controller architectures and renewed motivation to move to 48V power rails.

As designers of semi- and fully-autonomous vehicles systems rely increasingly on nondeterministic, artificial Intelligence-based control systems, traditional prototyping and physical testing are no longer feasible, nor are they sufficient, as the number of theoretical test scenarios is exploding beyond control. Simulation and test tools must evolve from physics-based modeling and analysis to behavior modeling, allowing new vehicle designs to be “driven” millions of virtual miles before a system can be validated and signed off.

Siemens PLM’s Acquisition Spree

Siemens PLM has a long history of providing product development and engineering software to the auto industry in industries that share many of the characteristics of automotive product lifecycle, such as heavy equipment and aerospace manufacturing. The company’s continued growth in these sectors puts pressure to develop new tools and methods that must be in lockstep the growing complexity in product development the industry is experiencing.

Over the past five years or so, Siemens PLM has made several well-placed acquisitions, targeting the more pressing areas in automotive product development: software development, electronics design, and simulation. Siemens acquired test and mechatronic simulation software from LMS, which also supports model-based systems engineering. Later it added Polarion’s application lifecycle management (ALM) development environment.

Two very recent acquisitions: Mentor Graphics and TASS, are particularly important and indicative of Siemens’s aspirations and intent to maintain a leading role that is responsive to the changing needs of the automotive industry.

Mentor Graphics gives Siemens additional tools to address the needs of designers of high-density embedded electronics through a rich array of circuit and PCB design tools as well as comprehensive simulation capabilities.

TASS, Siemens PLM’s most recent acquisition, adds to the already diverse portfolio of simulation software tools designed to support the fast-growing space of ADAS and autonomous vehicles.

The Road Ahead

Siemens PLM has arguably the broader and most diverse portfolio of engineering and product lifecycle software tools among its peers, supporting product lifecycle activities from requirements management, to design and engineering, to digital manufacturing planning and factory automation, and to service lifecycle management.

Siemens was not always articulate enough in telling the product lifecycle management story. The individual tools—the individual book chapters, if you will—where there, but when it was time to weave them into a cohesive PLM story, some chapters may have been neglected and others weren’t necessarily in the correct order. As a result, the promise of PLM and the digital thread connecting the different phases of the product lifecycle didn’t always come through well enough.

The rapid rate of acquisitions of functionally-rich companies, the continued evolution of product development methods such as model-based engineering, and the overall neck-breaking pace of new automotive development that calls for frequent reassessment of the portfolio, is undoubtedly a major headache for Siemens PLM strategists.

To be fair, the vision and capabilities of modern-day PLM portfolio can be both abstract and overwhelming. Practitioners that grew up on “old” development methods need to revisit deep-rooted concepts and reexamine decades-old workflows in order to take advantage of the promise of the rich PLM environment and be convinced they can use it effectively to realize its benefits.

Automakers must step up their ability to develop, test and deploy software-intensive vehicle systems. The need to have a cohesive and complete view of product lifecycle information will encourage broader and pervasive adoption of PLM-enabling technologies and processes throughout the extended enterprise.

Granted, the challenges facing OEMs and suppliers are more involved and far-reaching than can be resolved overnight simply by introducing new engineering software tools. Automakers must enhance their capacity and the capacity of the extended value chain for better product lifecycle decisions through product lifecycle management practices that improve visibility and agility throughout the product lifecycle. The new digital enterprise should broaden the use of PLM software and maintain a digital thread that connects value chain functions and engineering disciplines to create a high-integrity rich multidisciplinary decision-making context throughout all lifecycle phases.

Judging by the recent acquisitions, Siemens is focusing on the right growth areas. The company is also undertaking efforts to integrate user interfaces and workflows, improve application interoperability, and create a unified, seamless PLM environment to enable the digital enterprise built on the Teamcenter backbone.

This blog post was sponsored by Siemens PLM

Image: Driverless Car (Popular Mechanics)