Universities feeding automakers must get their students ready for new tools.
The electronic content of automotive systems is drastically increasing. This includes both hardware and software. In 2013, the Aberdeen group conducted a survey that highlighted embedded software development trends in automotive.
For example, 18 percent of the respondents indicated that the greater use of embedded software will be the most significant change to the automotive industry within the coming years.
When asked “What are you doing to achieve required fuel efficiency and emission standards?,” more than 40 percent of respondents at OEM and Tier 1 companies highlighted the greater reliance on embedded software to control systems.
The direction towards autonomous driving is putting even more emphasis on the use of embedded software.
IHS, a global information company, highlighted at an ADAS conference the following requirements as the key technology needs for autonomous driving: sensors, software to interpret sensor data, software and hardware enabling the control of the vehicle, and algorithms/software implementing the behavior of a skilled driver.
Although there are technical challenges associated with these developments, everyone believes that they can be solved, but at what cost?
Here, I would like to focus specifically on the brain power required to solve and implement these systems.
Since 2012, the automotive industry has seen an increasing demand for automotive engineers. An article from Bloomberg on August 14, 2012 stated that the shelf life of an auto engineer looking for a job in Michigan was three days; that additional pay and bonuses were required and that automotive companies were recruiting across the entire country.
In addition, we should notice that when it comes to electronic hardware and software, the automotive industry needs to compete with a broader range of industries that are often perceived as more attractive (and hip) to the younger crowds of engineers hitting the labor market.
The trends highlighted above were reflected in a recent presentation I attended from General Motors stating that the gap in development for Electric/Electromechanical (E/E) architecture and implementation simply cannot be filled with more human power.
It also needs to be filled with better development methodologies and tools. For example, today the software architecture in a vehicle handles more than 200 software tasks and 500 bus signals; the hardware topology cover more than 60 Electronic Control Units (ECU) and more than 10 serial data buses.
However, the use of a virtual development environment for E/E design in automotive companies has not followed the same pace as the use of a simulation environment for mechanical design and most of the work is still performed with physical systems.
Virtual hardware prototyping provides engineers the ability to simulate digital hardware for the purpose of developing, integrating and testing automotive system software. Such tasks as complex driver development, multicore software development, AUTOSAR software development, functional safety test development, validation, and execution can be performed more efficiently in the virtual hardware prototyping environments.
This approach has been deployed in other industries such as mobile and consumer, and is slowly making its way into automotive as described in “Better Software. Faster!,”a practical guide about the use of virtual prototyping tools for concurrent hardware/software development. (A free copy of the book can be downloaded at here.)
Accelerating the use and deployment of such methodologies and tools requires that young engineers are properly educated on their use.
Is the problem today that software development, integration, and test education remain focused on the enterprise or the more ‘sexy’ devices, such as mobile and consumer? Should universities and colleges enhance their curriculum to ensure that practical implementation and test tools are used? Should the automotive industry invest in their feeder schools to train young engineers on the tools that allow them to deliver automotive electronic and system software faster and more reliably?
I believe that everyone must step up to enable the next generation of engineers to hit the market prepared to adopt and deploy these new tools and methodologies, as well as realize the exciting challenges of automotive electronic and software design.
Without it, automotive will remain behind and be growth limited. As a technology supplier, Synopsys is ready to support universities and colleges serving the automotive industry. Are you?
-- Synopsys Automotive Seminar - OEM’s Virtual Development and Integration (VD & I) of In-Vehicle E/E Architectures – Paolo Giusto GM Presentation