And it’s not just with self-driving cars.

Oftentimes, when people discuss or hear about Artificial Intelligence (AI) in the context of vehicles, they jump forward to the topic of self-driving cars. I say “jump forward” because they’re overlooking the fact that AI has a much broader and foundation-disrupting impact across the entire automotive industry — from manufacturing and predictive maintenance to a slew of AI-enabled in-vehicle experiences, such as personalization, in-car virtual assistants, driver behavior monitoring, and intelligent driver assistance systems. In fact, these new “in-car experiences” are becoming the most important purchase criteria for consumers beyond your typical horsepower, handling and a good stereo system.

But autonomous car lovers should not give up — your dream is not far off. The fact is, AI in the near term should enhance in-vehicle experiences and pave the road to true, autonomous driving. Let’s go over how…

Redefining in-vehicle experiences

Currently, AI can personalize car experiences for drivers and passengers, facilitate contextual safety and new kinds of intuitive and natural interactions. For example:

> Personalized experiences: Your car could recognize you by your face, voice or fingerprint — distinguishing you from others and personalizing car settings, driving mode sportiness or comfort level, lighting and the infotainment system based on your preferences. The AI-based in-cabin camera could also recognize other passengers in the car and customize audio settings and content for each passenger to enjoy their favorite music or movies based on their preference without disturbing other passengers.

> Contextual Safety: Your car may be able to act as driving buddy through intelligent driving assistance systems, including in-cabin monitoring and ultra-HD surround view monitoring. For example, inward-facing AI cameras could recognize a person’s facial expressions, voice, gestures, body language and more, to ensure the driver is awake, alert and engaged. The intelligent driving system would make the driver more aware of his surroundings and driving conditions. If the car spots icy road ahead, it could alert the driver, slow down and automatically engage its all-wheel drive system.

> Intuitive and natural interactions: You may be interacting with your car through voice, gesture and the car would understand your needs, preferences and habits. For example, you can raise the music volume using gesture or ask the car for food recommendation on your way home through voice control and natural language understanding. The car can alert you that you need to charge your EV car and you can grab a bite on your way to the EV station. The kids are getting bored in the backseat, the car might even recommend a movie and ask you if you want to play it for them.


Paving the road to autonomous driving

The SAE and NHSTA define six levels of autonomy from Level 0, meaning no driving automation, to Level 5, meaning full autonomous driving. Only Level 5 means no driver involvement, while all other levels require some driver input or intervention, which is called “hand-off.”

Using AI for in-vehicle experiences will play a key role in the transition to full autonomy, especially Levels 3 and 4:

> Provides confidence in automated vehicles: Through the AI-enabled cockpit, the vehicle will inform the driver of what it sees on the road and how it is handling different situations, so the driver can trust the automated driving system.

> Eases the hand-off process for safer driving: The decision of when a “hand-off” should take place to pass control between a vehicle and a driver is a hugely important, life or death, issue for Level 3 and Level 4 vehicles. AI-based driver monitoring will help ensure that driver is alert and awake to be ready to take over. Also, AI can help in alerting the driver when the vehicle encounters a situation that requires her intervention.

Magee Zaki - Director, Technical Marketing, Qualcomm Technologies