Through 2030, analysts predict impressive annual growth rates between 15 and 30 percent for the shared vehicle market. OEMs will be forced to ask themselves “How can we offer a seamless brand experience with our vehicles to assure the loyalty of users that are not owners or are not in a position to buy?” and “How can we create a superior user experience when usership outweighs ownership?”
In a first article of this series, I talked about available technologies that can help personalize the in-vehicle user experience for shared mobility concepts. This time, I had the chance to talk to Michael Kaisser, Senior Principal Innovation Manager at Nuance, about use cases and capabilities of a smart car-sharing application as part of the voice-enabled automotive assistant. This application might include a variety of functionalities that will make the car sharing user experience easier and more convenient and could become a distinguishing factor for car manufacturers as well as mobility service providers.
Michael,what is the main idea of a car sharing app as a part of the voice-enabled automotive assistant?
Michael Kaisser: Today, many OEMs are operating their own shared mobility fleets or are offering their vehicles to third-party mobility service providers. With shared mobility and the new concept of using a car instead of owning it, user experience is becoming one of the most important distinguishing factors. This means both user experience of the provider’s offer and services but also in the vehicle. We know that the automotive assistant—both in the car and on a companion smart-phone app–adds a lot of value to the car sharing experience.
What needs to be done to realize the idea?
Michael Kaisser: The latest advancements in artificial intelligence already lead to a personalization of the driving experience. Vehicles – with the help of our automotive assistants – have access to world knowledge and contextual information, and they are constantly learning about user preferences with every interaction. Our hybrid approach with a strong focus on vehicle-embedded services is most valuable as it enables us to keep large amounts of functions and data embedded in the car while also utilizing the cloud for access to content and services. Shared mobility will also make use of this hybrid approach – using on-board systems to access vehicle sensors and data. Since drivers are using different cars, user profiles need to be shared in the cloud for further processing in every vehicle the user rents.
When we are talking about car sharing, we are not only talking about car-related aspects but also about a complete service infrastructure. How does this affect a car-sharing application and the automotive assistant?
Michael Kaisser: First, we need to keep in mind that the individual design of the car-sharing concept is different with every provider. Therefore, we choose the same approach for our shared mobility application as we choose with every automotive assistant development: we create a tailormade solution which fits the individual needs of the car manufacturer and/or mobility service provider. Car-sharing providers typically offer an app that enables users to rent cars and manage their accounts. Considering this, a car-sharing solution must offer seamless interoperability between the automotive assistant and, for example, a mobile device that’s running the provider’s app.
Let’s have a look at the end-to-end user experience, i.e. at the user experience from starting the rental to returning the car. As with every Nuance application, voice – and more importantly natural language understanding – will play a major role here, right?
Michael Kaisser: Indeed. Natural language understanding enables the most natural start of the car rental possible. The user might ask his car-sharing application questions such as “Is there a free car around here?” or “Please notify me when a minivan becomes available within 500m of my current destination.” NLU makes sure that the system is understanding the request and knows how to further process it. Leveraging artificial intelligence about the user’s preferences, we could even go one step further and have the user search for his “favorite vehicle” or the “same model as last time.”
Alright, let us assume the system found a car nearby, it is reserved, and our user is starting the rental. Which additional features can the smart automotive assistant application offer?
Michael Kaisser: Users who are new to a particular car sharing provider may not be familiar with the process of reserving a car. It makes sense to offer a smart, interactive car manual that gives a first introduction to the most important features and functionalities. Combining this function with personalized user profiles that are safely stored in the cloud and made available to the system, we can make sure that the user gets his individual introduction, that takes the rental history into account. If the user has already driven the car brand or the vehicle model in the past, the system might omit information or show the differences to the last driven vehicle only. In addition, the smart car manual enables the user to simply ask the system a question whenever he encounters a situation he is not used to.
Furthermore, the system could proactively make recommendations or provide useful information such as the position of the tank lid when approaching a gas station or a reminder to recharge a fully electric vehicle.
What if something goes wrong and the car won’t start, a failure is detected etc.?
Michael Kaisser: Service providers are investing significant amounts of money to assure customer satisfaction. If a car won’t start, several scenarios are possible – both with respect to the problem as well as the solution: Sometimes, it’s only a small problem that prevents a car from starting – e.g. the user did not push the brake as he is not used to driving a car with automatic transmission, or the door is not completely shut. There are many pitfalls – especially if users are not that used to a specific car. Possible solutions could include onboard diagnostics, a dialogue with the electronic car manual, or routing the request to a human operator for further processing.
The last step of the customer journey is handing back the vehicle. What is the role of the car-sharing application in this process?
Michael Kaisser: Comparable to starting a rental, terminating a rental requires a few steps to be performed in the correct order. In addition, sometimes users want to park the car but keep it reserved, perhaps because they left their shopping bags inside the trunk, for example. Using language for this process is natural and convenient for the users, since you can just say what you want and do not have to fiddle with the app. Another possible scenario is an expiring rental period. If the user is still driving, maybe because of a traffic delay, sending him a text message or an email warning about the end of the reservation can be very distracting. Using voice, the automotive assistant could have a simple quick conversation with the driver about extending the reservation.
Michael, thank you very much for your time