Risk Management for Embedded Software Development

June 13, 2017

Formula One Race Car is fast like development needs to be for ADAS

What do successful downhill skiers, Formula One racers, and mountain climbers have in common? They learn as much as they can about their sport and they use the very best equipment available. By taking this approach, they transform pure risk into calculated risk with tangible payback. But developing embedded software isn’t a dangerous sport, you say…

Road to safety graphic

In truth, embedded software development for advanced driver assistance systems (ADAS) and autonomous driving applications may be riskier than you think – at least it can be unless you, like those flying down a mountain at nearly 100 mph or ascending an exposed cliff face, have accumulated the knowledge and tools necessary to reduce risk.

What Could Go Wrong?

ADAS and autonomous driving technologies rely on embedded solutions – a combination of embedded software, microprocessors, and printed circuit boards (PCBs) – to enable features such as adaptive cruise control, lane departure warnings, automatic shifting, etc. We’ve all seen malfunctioning electronics – think “Microsoft Windows® Blue Screen of Death.” But when a car is traveling 70 mph and suddenly shifts into reverse, or a sensor miscalculates the distance of an object, or the brakes don’t work as expected, catastrophic injuries or even death can result.

Even less dramatic malfunctions of embedded software, such as electronic emissions control, can lead to significant monetary loss and substantial reputational damage. (Just ask Volkswagen, whose stock fell 25% after their “EmissionsGate” scandal.)

What’s an Embedded Software Company to Do?

Despite the business risks associated with developing embedded software for ADAS and autonomous driving applications, no automotive embedded software company can afford the equally looming danger of being left behind in the automotive industry’s hottest market. With an estimated compound annual growth rate (CAGR) of 16% to 18% through 2020, the global ADAS market represents huge revenue potential. Likewise, IHS Automotive forecasts the autonomous vehicle market to reach a 43% CAGR between 2025 and 2035.

Here are a few ideas to help reduce business risk for developing ADAS and autonomous driving embedded software:

Learn about safety standards and integrate them into your development cycle
Use safety-certified hardware
Chose an ASPICE® Level 2-certified compiler, debugger, and linker that can exploit hardware safety and performance features
Design embedded solutions to reduce the risk of driver distraction
Incorporate monitoring and redundancy into your embedded software design to detect errors in real time
Take advantage of cryptography and performance libraries to build secure, high-performance code
Make sure you can perform RTOS (real time operating system) configuration and debugging for time-critical embedded applications
Plan for over-the-air (OTA) updates of your embedded software, using position-independent executables

Formula one car again

When You Have Confidence, You Can Move Fast

The Formula One racer knows and trusts his car, and knows the track. He can go 200+ mph without undue risk. Similarly, by choosing the best hardware and embedded software development tools, adhering to automotive safety standards, and integrating safety and security into your embedded software designs, you can accelerate your embedded software company into the lead in the race to design the best ADAS and autonomous driving applications.

Learn more by reading the white paper, “Taking the Business Risk Out of ADAS Development ”.