The dynamics of a dynamic HMI
Andrew Thompson and Jan Stojaspal report on how situation awareness can make for a more personalized, less distracting human-machine interface (HMI).
An estimated 421,000 people were injured in car crashes involving a distracted driver in the United States in 2012. Texting while driving made an accident 23 times more likely to happen. And 21% of 15- to 19-year-old drivers involved in a fatal car crash were distracted by the use of a cell phone.
The statistics go on and on, and they all end in hazard. You could, for course, try to force people to leave their connected devices – be they smartphones, tablets or in-vehicle infotainment units – alone through admonishment and legislation as the U.S. government has been trying to do for at least half a decade.
In 2009, President Barack Obama issued an executive order prohibiting federal employees from texting while driving on government business or with government equipment. A year later, the Federal Motor Carrier Safety Administration banned commercial vehicle drivers from texting while driving. And a variety of U.S. states have since issued a patchwork of rulings banning everything from texting while driving to Google Glass.
But the lure of connected devices can be irresistible. According last year’s survey by the National Highway Traffic Safety Administration, some 660,000 drivers are using cell phones or manipulating electronic devices while driving at any given daylight moment in the United States.
There is another way
Luckily, there are other ways to mitigate driver distraction. And one of them is through what is known as a dynamic or situation-aware HMI. Unlike a regular HMI, which is static in what information it presents and how it presents it, a dynamic HMI adjusts itself to the driver’s situation – his route, passengers, weather and so on – resulting in a less distracting, safer ride.
For example, a car with a dynamic HMI might suspend all calls or texts if dangerous weather or traffic conditions were detected, and, as those conditions improved, present only the most important calls and texts. Or it would be able to determine who is in the vehicle and choose the type of music or filter incoming calls. For example, if your boss were in the car, the HMI might decide to skip over death metal on Pandora and only route work-related calls to the vehicle.
“The technology is there,” says David Smith, assistant professor of automotive engineering at Clemson University in South Carolina. “One part is getting the cost down and the other is pure design work of OEMs taking them and putting them into a full-fledged consumer product.”
The next arms race
In many ways, HMI is emerging as one of the biggest differentiators and brand identifiers for OEMs. Increasing the situational awareness of the car provides even more avenues for brand polishing and impressing consumers. “I think that the next arms race in the automotive industry is defining these dynamic and interactive experiences that are context-aware,” Smith says.
And there are plenty of examples of carmakers going in just this direction.
Take Mitsubishi’s recently announced “Ultra-simple HMI” that predicts a driver's needs for the safe and easy operation of vehicle equipment in one or two steps and no more than 15 seconds. The concept bases its predictions on operational history (past destinations and routes, searches and in-car device usage) and current conditions (location, speed, fuel level, current traffic, school zones and other road conditions) to minimize the steps and time required to operate the vehicle's navigation, phone, air conditioner and audio-visual equipment.
Audi is another carmaker looking hard at situation-aware HMI. Its Audi Urban Intelligent Assist research project, conducted in conjunction with the University of Southern California; University of California, San Diego; University of California, Berkeley PATH and University of Michigan Transportation Research Institute, is looking at Big Data, attention monitoring and the ability to learn from a driver’s schedules, habits and preferences, to deliver a highly customized driving experience.
But modeling things in a lab and pulling them off as commercials products are two different things.
“Currently, even the most advanced HMIs on the market are not very dynamic in nature,” says Vasily Suvorov, VP technology strategy, Luxoft. “There seems to be an effort to improve personalization and integrate various advanced driver assistant system features into the HMI, but still, there is very little built-in ability in current HMIs to automatically change in response to a driving situation.”
According to Suvorov, what needs to be at the heart of a dynamic HMI is a system capable of gauging the driver’s workload by analyzing feedback from various car sensors, including speed, how gas pedal and brakes are used, position of the steering wheel and real-time traffic information. “I envision, for example, HMIs that will understand the driver's schedule, correlate that with real-time traffic situation, driver preferences and offer an optimal route,” he says.
Once the driver’s workload is well understood, “we can make decisions as to what information, user interface … need to be presented to the driver at any particular moment in time,” Suvorov says. The busier the driver gets, the more simplified his navigation map may get, for example.
Another potentially important feature is collaboration between the driver and passengers under particularly difficult driving conditions. “For example, if I am driving and the road condition requires most of my attention,” Suvorov says, “then my passenger with a connected mobile device can assist me with navigation or answering calls.”
(For more from Suvorov, see Video: Luxoft on mitigating driver distraction through situation-aware HMI.)
Tesla’s Model S provides one of the best examples of where a dynamic HMI might head. Beyond the glitz of its 17-inch touchscreen, the vehicle’s navigation system combines Cloud and local data to improve navigation. But, for now, that’s about it. And it will likely take a few more years until HMIs actually integrate the features forecasted by experts.
Still, if the technology is there – or at least almost there – why wait so long?
One reason is that developing the component technology takes a lot less time to develop than a car itself. “Six to 18 months is an eternity in terms of consumer electronics development, but less than one fifth of the time for an average car development program,” says Telematics Update’s Automotive HMI Report 2013.
Andrew Thompson is a regular contributor to TU. Jan Stojaspal is TU’s executive editor.
For all the latest telematics trends, check out Telematics for Fleet Management Europe 2014 on March 12-13 in Amsterdam, The Netherlands, Content and Apps for Automotive Europe 2014 on April 8-9 in Munich, Germany, Insurance Telematics Europe 2014 on May 6-7 in London, Telematics India and South Asia 2014 on May 28-29 in Bangalore, India, Insurance Telematics Canada 2014 on May 28-29 in Toronto, Telematics Detroit 2014 on June 4-5 in Novi, Michigan, Advanced Automotive Safety USA 2014 on July 8-9 in Novi, Michigan, and Telematics Munich 2014 on Nov. 10-11 in Munich, Germany.
For exclusive telematics business analysis and insight, check out TU’s reports: Telematics Connectivity Strategies Report 2013, The Automotive HMI Report 2013, Insurance Telematics Report 2013 and Fleet & Asset Management Report 2012.