Time to take fresh look at test tools and approaches for automotives

While automotive makers exploring vehicles of tomorrow with advanced control systems like connected car, ADAS and autonomous vehicles there is a need to relook into the test tools. Bob O’Donnell drives through the missing connection between the automotives and software.

Industries continue to evolve and several sectors across the board have seen varying levels of disruption. Among these sectors, the automotive space has been given immense focus, with elements such as fuel economy, emission control and improved driver control being the key concerns. Each of these areas, in the last five years, has benefited from the influx of innovative technology into the automotive sector, resulting in rapid progress in Electronic Vehicle (EV) technology, autonomous vehicles, ADAS and many more. India in particular has put in place multiple initiatives geared towards encouraging shared and electric mobility, with subjects such as powertrain, vehicle concepts and transmission being consistently researched upon. In fact, next year is going to be an important one for India as multiple electric vehicles are set to launch in the market.

As we inch closer to 2020, this collaboration between automotive and technology is only deepening and the efforts taken by automotive companies to adopt and invest in new technologies is becoming clearer than ever. Be it for building safer cars or making them smarter, the convergence of auto and tech is making it not just possible, but also extremely easy for the consumer to buy and own some of the most advanced technological innovations. However, there are several challenges that still exist in the convergence process- primarily lying in the production and testing of key technology-based hardware components and system software. While the industry as a whole has shown multiple trends that hint at successful integration between the two domains, many auto vendors are still utilising outdated processes that are no longer compatible with and don’t meet the demands of technologically advanced components.

In-vehicle systems get smarter

Recent trade shows across the globe have witnessed the rise of one vehicle electri?cation as a major trend. New car models have more semiconductors and electronic components that are control bigger portions of the vehicle’s capabilities. This has underlined the vast increase in the power of the chips being installed; vendors like Nvidia, Qualcomm, Intel, and ARM—primarily designing chips for smartphones and computers, are now standing alongside automotive suppliers like Renesas and NXP in the process of creating powerful car computers. These computers have moved beyond merely supporting in-vehicle infotainment (IVI) systems to now powering more Advanced Driver Assistance Systems (ADAS) and even sensor fusion. In the recent past, these advances have allowed companies like LG and Dyson to experiment with electric powertrains, while others like Panasonic and Harman focus on advanced in-car digital cockpit experiences.

Two worlds in need of integration

It seems safe to presume that automotive ECUs demand the same kind of test procedures used on products like smartphones and other mobile devices. However, that is not true at all, mainly due to the legacy nature of the automotive business where mechanical approaches still take precedence to a large extent. Automotive testing has always been complicated due to the nature of electromechanical systems present in cars; and modern day vehicles pose further challenges in their test, validation and simulation requirements, given the software-based nature of their key systems and the increasing complexity of interactions across multiple digital components.

In the current industry scenario, mismatch between advanced components and traditional testing methods have led to several roadblocks to a quicker time to market. Many tier-1 automotive suppliers and car makers still adhere to one-at-a-time testing of ECUs instead of adopting a parallel test-methodology. Additionally, there is also a lack of required level of test suite sophistication that is needed to detect flaws in a code-heavy ecosystem as well as the inability to adapt to adapt to new software builds. The challenge moving forward, hence, would be to integrate the worlds of automotive and technology from a test and methodology perspective and develop new approaches that build on the automotive industry’s core focus of reliability and safety testing. This focus has gained additional importance in the recent past, courtesy the development of autonomous vehicle technology that take important driving decisions on behalf of the driver. This brings in the need for companies to extensively test these capabilities in a cost-effective manner to ensure that all flaws within the system are eliminated.

What does this mean for test engineers?

At the test level, companies must now look to create parallel test architectures, automate tests, design products and services with software upgradability in mind, and validate interactions with new technology ecosystems, among others. This involves the use of test tools that are both sophisticated and industry-proven across challenging environments for optimal results. These test tools must also look to create innovative test environments that utilise a more ?exible, software-based approach that allows modular hardware and software components to be integrated dynamically. Companies like NI possess the required capabilities to be effective in such a scenario; through the use of a ?exible platform-based approach to create a customisable test system IP, these companies that place the power to adapt in the hands of automotive test teams rather than third-party vendors.

As automotive manufacturers evolve from carmakers to mobility providers, it’s time for them to take a fresh look at their test tools and approaches. They need to overcome the pressure of new and rapidly changing test requirements using an open and easily upgradable platform that is designed for test system ?exibility, and that enables them to test the vehicles of tomorrow today.

Bob O’Donnell is the president, founder and chief analyst at TECHnalysis Research, has a multi-faceted career in the technology business. He is widely regarded as an expert in the technology market research field and his original research and advice is used by executives in large technology firms all over the world. He is a prolific author and content creator as well, having published 339 articles and podcasts.