On the road to autonomous vehicles

Founded as a spin-off from Motorola in 1999, ON Semiconductor has been active since 2000. Today, the company employs approximately 30,000 people worldwide. Saleswise, automotive is the largest sector for ON Semiconductor, representing 30% of total company revenue.

ON Semiconductor is a key player within the ADAS and autonomous technology market. Its position allows it to understand what technical and other hurdles must be successfully tackled in the future to reach the next levels of automation.

Somo firstly highlighted the key automotive megatrends: autonomous vehicles (AVs), connected vehicles, and vehicle electrification. With regards to AVs, the implications surrounding them include safety, improved efficiency, inclusion and access benefits, and new paradigms – the ride sharing economy. The key technologies are sensors, radar, LiDAR, GPS, 3D mapping, deep learning and artificial intelligence.

The second area, connected vehicles, refers to vehicle-to-vehicle or vehicle-to-grid communication. Connectivity is becoming increasingly important within the vehicle, both in regards to infotainment and collecting data from sensors in order to assess the driving environment.

Finally, Somo discussed vehicle electrification. In many cases, car companies are looking at vehicle electrification and AVs together. Vehicle electrification does not only refer to stop start engines, it refers to the increase of motors. “I don’t think many of us can remember the last time we rolled down our windows or used a manual lock,” said Somo.

What is encouraging this ‘drive’ for AVs? Firstly, AVs reduce traffic incidents by eliminating human error. Somo explained that in the US there are approximately 40,000 fatalities per year, of which 94% are attributed to human error. On a global scale, this number rose to the hundreds of thousands. There is therefore substantial opportunity to reduce the human error element and make our roads safer.

Driverless vehicles also have efficiency benefits. They improve traffic flow – you can fit more vehicles onto our roads. We can therefore get more efficiency out of the infrastructure we have already in place.

There are also societal benefits, such as inclusion and access. Those with a disability who previously couldn’t access a vehicle can make use of the constantly improving ride sharing services.

Are we ready for this autonomy? The car industry is currently situated between Level 2 (partial) and Level 3 (conditional) automation. In the former, the vehicle has combined automated functions, like acceleration and steering, but the driver must remain engaged with the driving task and monitor the environment at all times. In the latter, the driver is a necessity but is not required to monitor the environment. The driver must be ready to take control of the vehicle at all times with notice.

In order to progress to Level 3 conditional autonomy, companies must consider when is a reasonable time to expect a driver to regain control of the vehicle to avoid any obstacles when notified by the car that it is unable to continue navigation. According to Somo, “that is a debate that is making it difficult to roll-out Level 3 systems and may prompt some companies to skip over this and actually go straight to Level 4 high automation.”

In Level 4 high automation, the vehicle is capable of performing all driving functions under certain conditions, while the driver may have the option to control the vehicle. Companies prefer this because they can clearly define an operating domain for the vehicle where it can always perform the functions and monitor the driving environment, and then make a seamless transition back to the driver over a reasonable amount of time.

In conclusion, transportation is clearly at the brink of disruptive change. Automotive companies, technology companies, universities and governments must collaborate to enable safe, secure and timely development of AVs.