09 Aprile 2008


Industry is working at cars that will talk more and more to their drivers and some researchers are even working at car interfaces that automatically adapt to the specific user who is driving. I’ve asked some questions about it to David Krum, project manager at Bosch Research. David is one of the organizers of the special interest group “Interaction in the Automobile” which meets this morning at CHI 2008. You can read his interesting answers in the following:

Thanks to computers, the number of functions that cars are offering to drivers is increasing. Is this always for good?
“Many functions which are made possible through computers, like anti-lock brakes and electronic stability control have made driving much safer.  Other functions, like driver navigation systems and multimedia players make our trips more convenient and more pleasant.  However, there is a need to integrate such features into a seamless experience that won’t distract or overwhelm the driver.  It can become hard for drivers to find particular multimedia functions if they have to navigate through a lot of menus.  A map display may have too much information on it, so the driver will stare at the display for a few critical moments too long, not looking at the road.  Our goal at the conference is to make sure that the fields of human-computer interaction and automotive design come together.  We need to work harder to determine the best way to present all the new functions that customers want.  We also want to find better ways to inform the driver of hazards and let them know what their car and other cars are doing.”

How the automotive industry is approaching the design of in-car interfaces?
“The automotive industry uses a wide set of approaches to design in-car interfaces.  Some, like design guidelines and design reviews, are used during the design process to make sure we are building a good interface.  There are also rounds of acceptance testing and user testing which are applied to products, once they are built or prototyped.  Of course, we also listen to customer feedback and reviews from industry writers.”

What is the NHTSA 100 Car Naturalistic Study and is it important for interaction design?
“The NHTSA 100 Car Naturalistic Study is important since it was a long term study that equipped cars with the automotive equivalent of an airplane’s ‘Black Box’.  By recording videos from inside and outside, as well as radar, GPS, and other sensors, the investigators could see how people drive and interact with devices in the car.  During the two million miles of driving during the study, they did record some accidents and near accidents and they used the recorded data to determine the causes.  In many cases, interacting with devices contributed to unsafe driving.  This is a deep source of data that we will examine closely to learn how to make interaction unnecessary or at least less distracting.”

Could you give me a few examples of particularly interesting work that will be presented at your special interest group?
“We have invited representatives from the automotive industry and students and professors from universities to talk about the interaction issues we face and possible solutions that are being researched. Some researchers will present guidelines for the design of speech interaction systems.  It has been shown that it is important for cars to know their drivers so the car can tailor its voice and responses to the emotional state, rhythms, and current workload of the driver.  If the cars can do this, the driver will be less distracted and will actually drive better. Others will talk about software that can be used to mathematically model driver performance and how driving simulators can be used to test how distracting devices can be.”

How do you imagine the interaction between driver and car in the future, thanks to the research that is being discussed at your special interest group?
“We will definitely see near term advances in wireless connections, interconnections between mobile devices and the car, and increased use of speech interaction.  We expect increasing use of Internet connections in the car, for dynamic navigation information, streaming music, and news.  Customers now expect the car to connect and integrate portable devices that they bring into the car, like MP3 players, cell phones, and even laptop computers.  Finally, automotive makers are starting to look at methods of monitoring driver state (Is the driver falling asleep?  Is the driver distracted?) in order to determine optimal HMIs for in-vehicle systems.  All of these novel use cases present new challenges in interaction design. Some longer term changes we may see could include head up displays, which can project information on the windshield and highlight objects around us.  We will eventually see vehicles become increasingly autonomous, so they may be able to drive themselves in certain conditions.  There will still be interaction issues with semi- and even fully autonomous vehicles.  How will the car signal that it needs assistance?  How will the car transfer control to the driver?  How will the car gain the trust from the passengers?”

If you found this interview interesting, do not miss the following one. It will give you the perspective of Brian Lathrop (Volkswagen Electronic Research Lab).

© 2008, Il Sole 24 Ore. Web report from CHI 2008.