Gerhard Nuessle

Gerhard Nuessle

Most carmakers are working on ways in which to weave in some level of Artificial Intelligence (AI) and voice assistance into their model range. The Mercedes-Benz new A-Class incorporates the Mercedes-Benz User Experience (MBUX), the carmaker's latest interactive telematics platform. This intelligent, voice-controlled infotainment system - co-developed with Samsung-Harman - is characterised by its ability to learn using AI. To learn more, we caught up with Harman's Senior Vice-President of Operations Daimler, Connected Car Division, Gerhard Nuessle. In this interview, he explains the intricacies of the MBUX system and why he believes it will revolutionise the UX.

A wholly-owned subsidiary of Samsung Electronics, Harman was among those recognised at this year's Daimler Supplier Awards for the role it played in developing the MBUX. Last month, Harman said it will introduce augmented reality navigation as part of the MBUX, marking its first ever use.

Can you give us a general overview of the MBUX system for Daimler?

Yes of course. The MBUX system is the latest Daimler infotainment system and debuts in the new A Class that's now in production. It is based upon one of the most powerful SoCs (System on a Chip), offering truly state of the art computing power. In the car that means it offers incredible ability for the system to learn, to create a more personal user experience. It is also an infotainment system that is always up to date, thanks to Over the Air (OTA) updates. We think it is the most powerful infotainment system on the market and the feedback from the media that have experienced the system is very positive.

You say that it is the industry's most powerful infotainment system. What makes you believe it to be so powerful?

I think that is down to many factors but primarily that it is incredibly intelligent. MBUX learns your preferences and behaviours. For instance, it will start suggesting music based what you like and previous choices. Another example is how it learns your routines. If you leave the house at 7am every day to head to work, MBUX will automatically plot your route, giving you the latest traffic information and showing you the quickest way. The AI system embedded within the system is responsible for all of this. It is constantly seeking to understand the end users' behaviour and customise to their preferences or routines. It's a system that is incredibly intuitive and I think it will be a step change for users As Daimler stated at the launch 'Intelligence makes it individualised'.

It offers OEMs a real chance for an emotional connection to be created between the car and driver.

One other point I think that it is really interesting; for OEMs it offers a real chance for an emotional connection to be created between the car and driver. MBUX will be a powerful way for Daimler to connect with their customers in a way that just isn't possible today.

There are also other design factors that make us believe this is the most powerful unit: how smooth the operation of the platform is and how seamless the transition is between the different elements of the system. Performance wise, we think it is unrivalled; combining the graphics power, the audio capability, the augmented reality technology, as well as the natural speech recognition.

You mention speech recognition, is that how users can control and interact with the system?

Yes but the control is multi-modal, so the end user can operate the system either by voice or touch. The voice activation is driven by a very powerful natural language system that is both fast and smooth. Conversely, if a customer prefers to use touch control, then they can, using the display. Today's consumers have high expectations from their electronic devices in terms of graphics and interface. We have worked hard to ensure MBUX meets those expectations. You can touch, pinch to zoom, touch or swipe for pinpoint control. There are multiple ways to control functions; touch screen, touch pad, steering wheel letting you choose the most intuitive. But going back to your question, the voice control is very effective not just with literal commands such as 'play track seven' but it also understands that if you say, 'I'm too cold', it will raise the temperature.  

That's a big step from many systems today. How was this achieved?

The Cloud provides a far greater recognition accuracy, resulting in a more satisfying UX.

From a technical point of view the speech engine is embedded in the system and runs in parallel with The Cloud. With cloud computing services you are not limited by hardware capability. When an end user speaks, the voice recognition software cross-references the voice command with a huge database and compares. It then selects the command that scored the highest match and carries it out. The Cloud provides a far greater recognition accuracy, resulting in a more satisfying UX.

You say that the system is constantly being kept up-to-date via OTA updates, how does this work and how are the potential threats of this being managed?

OTA updates are provided by our technology and it is integrated into the system through the telematics. Any software updates run in the background making it seamless for the driver. Once updates are ready, they require the end user to approve. In terms of how we are managing any potential security risks, as mentioned, any updates are made through a connection with the telematics module, connected to the Daimler Cloud, so risks are mitigated through security features linked to that.

It is also worth noting that the telematics unit also can update other components in the car, not just the infotainment system. The technology is there, it is just a case of whether the component capacity is there. This will ensure that going forward the system will remain future-proof.

You've said how the system will have augmented reality capability, how will this work?

The augmented reality capabilities are an extension of the navigation system. We use the external cameras of the car to give end users assistance in certain traffic situations, with information displayed on the screen. This is especially useful at complex traffic situations like crossing or junctions.

Let me give a real-world example. If you approach a set of traffic lights and you are the first car waiting, it can sometimes be difficult to see the lights. You end up leaning over the dash. The MBUX system uses the external cameras to show the traffic lights on the display of your infotainment system. You will be able to see when the lights change to green.

Another useful example of the system's augmented reality capabilities is when navigating to a specific building. For example, you are heading to a particular house in a long street. The car's system will overlay a house number on the video picture displayed on the touchscreen display.

Typically carmakers roll out new infotainment systems on executive cars not smaller cars. Was this surprising for Harman?

No we aren't surprised. The target market for A Class customers is a demographic that is really interested in this type of technology and being connected. We are seeing across many of our customers that the infotainment system is of more interest than engine power or performance.