A study by pymnts.com suggests that improved connectivity and autonomy in vehicles will one day create a US$230bn commuter commerce market. Passengers and riders will be able to participate in a brand new ‘dashboard economy’, with autonomous driving allowing occupants to focus on their shopping, facilitated by in-vehicle human machine interfaces (HMI) as well as existing smart devices.
It’s a vision that underlines just how radically vehicle interiors and functions will change with the onset of CASE mobility: even today, shopping is not an activity associated with trips in the car. Like many other aspects in the connected, autonomous field, artificial intelligence (AI) will play a role. Natural speech recognition, for example, could allow people to navigate online stores and make choices via voice. AI development by automotive organisations has thus far focused on the outside of the vehicle—computer vision, path prediction—but as automakers shift their attention away from providing the best ride to the best in-vehicle experience, could AI prove the ultimate value driver?
Smart ride
“Cabins are going to smarter,” said Abdelraham Mahmoud, Senior Product Manager, Affectiva. Speaking at M:bility | California, he explained, “sensors like cameras and microphones can check whether a driver is paying attention to the road, but could also determine a person’s emotional state and respond accordingly.” For example, he says, a vehicle which understands an occupant is frustrated may look to correct its own procedures, or ready advanced driver assistance systems functions.
In short, cars could become experiences as immersive as any social media platform. Targeted advertising from services could be based on mood along with preferences. Affectiva is a company focusing on automotive AI for service features, as well as safety applications for automakers. “If the vehicle is driving itself, it becomes a matter of what services are being provided in a vehicle,” said Mahmoud, “and how to monetise these services.” One of the applications which Affectiva is focused on is measuring reactions to new advertisements for companies. “Imagine that the car will become a platform to test content and tailor it based on emotional response, as well as the context of the person in the vehicle.”
I think there will be much bigger demand for personally-owned vehicles with levels of autonomy than some people are predicting
If a vehicle can intelligently provide best-class infotainment, tailored to occupant tastes and context, that vehicle stands a great chance of succeeding in the car market of tomorrow. Indeed, this could prove so valuable to customers that Forrest Iandola, Chief Executive of AI developer DeepScale, believes the much-predicted abandonment of the privately owned vehicle is exaggerated.
“I think there will be much bigger demand for personally-owned vehicles with levels of autonomy than some people are predicting,” he said. “Consumers will buy cars where the most innovative things about it will be almost everything to do with the software: we’ve seen what happens with cell phones, where people tend to pay more for the phone service itself, as well as applications.”
Automakers, he predicted, could end up spending more money on building and updating so-called ‘autonomy suites’, with connected features and their own distinct eco-system, rather than the hardware in the car itself: “The consumer might be extracting more value from a differentiated autonomy suite, which is to say the software, than they do from hardware.”
Of course, there will also be opportunities for monetisation via AI technology gathering data outside of the vehicle, something which is already happening today. “OEMs are gathering huge amounts of data from the vehicle’s environment. That data can be packaged and sold to cities,” said Bola Adegbulu, Chief Executive and Co-found, Predina Technologies. “One application we see is pothole detection. They already have the data, so in some sense it’s a no-brainer.”
AI-driven computer vision has the potential to recognise a large number of features on the road which could prove of use to transport authorities and other bodies. Harsh-braking data is another example, which could highlight accident blackspots in built-up urban areas.
