09 Aprile 2008

DRIVER DISTRACTION. Interview with B. Lathrop (VOLKSWAGEN)

What human factors issues should future car interfaces take into account? The need to monitor the driver and to mitigate driver distraction and inattention came out often in talking with Brian Lathrop (Volkswagen Electronic Research Lab) Brian is one of the organizers of the special interest group “Interaction in the Automobile” which is meeting this morning at CHI 2008. Here’s the interview:

Thanks to computers, the number of functions that cars are offering to drivers is increasing. Is this always for good?
“This will inevitably place more cognitive, motor, and visual resource demands on the driver, leading to increased opportunities for driver distraction – and possibly increased accident rates. In order to counteract these additional demands it will be important to develop appropriate distraction mitigation strategies in the form of variable driver assistance systems. Such strategies will need to take into account the nature and severity of the distraction and deploy the appropriate warnings and countermeasures. Additionally, designing the appropriate HMI to mediate interaction between the driver and in-vehicle systems will be equally important. Using internal and external vehicle sensors can help to inform whether in-vehicle systems should be operated in the current driving conditions and circumstances.”

How the automotive industry is approaching the design of in-car interfaces?
“The automotive industry uses a number of approaches to design in-vehicle interfaces. One common method for designing and evaluating driver assistance HMIs is to prototype the system in a driving simulator and test its effectiveness with users. A common approach for designing and evaluating HMIs for infotainment systems is to perform usability studies with prototypes of the system (e.g., A-Samples, B-Samples, etc). In addition to these two common methodologies, heuristic evaluation, design guides, and market reviews (e.g., via JD Power) are applied to the design of the infotainment and driver assistance systems.”


What is the NHTSA 100 Car Naturalistic Study and is it important for interaction design?
“The National Highway Traffic Safety Administration has released a summary of the Virginia Tech Transportation Institute study of pre-crash naturalistic driving data collected through instrumented vehicles. Drivers used these vehicles in their normal daily routines, were given no special instructions, no experimenter was present, and the data collection instrumentation was unobtrusive. According to the report, nearly 80 percent of all crashes and 65 percent of all near-crashes involved driver inattention, just prior (within 3 seconds) to the onset of the incident.
This is important for interaction design because it suggests that important design considerations for future driver assistance systems should be the inclusion of a driver monitoring system. By monitoring driver eye gaze and/or head position/orientation (proxy for driver inattention), appropriate HMI warning types and strategies can be devised for purposes of mitigating driver distraction. Additionally, these systems can serve a double-use for certain types of user interactions with in-vehicle speech and manual interfaces (e.g., gaze-dependent).”

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?
“The trend in autonomy will increase, leaving the driver the opportunity to engage in an increasing number of non-driving activities (e.g., work, entertainment, communication, etc). It is unlikely, however, that the driver will be completely removed from all driving related tasks. Responsibilities will likely change from full control to system and state monitoring. This is similar to the way pilots’ responsibilities have changed in the aviation industry with the emergence of automated flight. The HMI will still be very important for safe travel, as strategies will need to be devised to ensure that the driver is rapidly and accurately brought back into the driving loop when critical events arise (e.g., system failure, decision points, etc).”

If you found this interview interesting, do not forget to read the one with David Krum (Bosch Research).

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