ADAS systems may not be the miracle solution to crash avoidance with one report indicating human behaviour is still a key variable.
A US study found while some systems like lane departure warnings could avoid collisions, the automated warnings could also risk making drivers complacent.
The research comes after an Australian insurance study found one in five drivers are turning warnings off because they find them too annoying or unnecessary.
Reducing Risk
Texas McCombs professor information, risk, and operations management Ashish Agarwal says the research indicates both results occur and tech designers need to adapt warnings to driver behaviour to better reduce risk.
“We compared cars with blind spot detection with the ones that don’t have that to see how it influences speeding and hard braking behaviour,” Agarwal says.
Controlling for variables such as vehicle features and driver demographics, the team found the two kinds of warning signals had opposite effects on driving behaviour.
Blind spot detection reduced the daily number of hard braking events 6.76 per cent and speeding events 9.34 per cent compared with cars without ADAS.
The research gauged two kinds of hazardous driving behaviour: rates of hard braking and rates of speeding. They also zeroed in on two sets of ADAS features that demand driver responses: blind spot detection and lane departure/forward collision warnings.
By contrast, lane departure/forward collision warnings led to 5.65 per cent more hard braking and 5.34 per cent more speeding.
These results grew worse over time as drivers grew accustomed to the systems the researchers found,.
Drivers speeded 0.40 per cent less often for each extra month they used blind spot detection.
They speeded 0.32 per cent more for every added month with lane departure warnings.
Two Kinds of Thinking
In a release Agarwal explains the warnings trigger two different modes of thinking.
Urgent warnings, such as lane departure and forward collision, require a driver to correct course immediately.
That causes System 1 or reactive thinking, which is automatic and largely unconscious, he says.
“It triggers risk compensation behaviour, which impedes your learning and makes your behaviour worse.” Says Agarwal.
By comparison, blind spot detection does not demand an instant reaction. It allows time for deliberate, System 2 thinking or reasoning one’s way to a response.
Future solutions
He believes OEMs should consider these findings as they design the next generation of ADAS features
“For example, a car could repeat a warning signal after a lane departure as well as during. That might encourage System 2 thinking, helping a driver reconsider their risky behaviour,” Agarwal says.
“For learning to take place, you need to be in the System 2 mode. That means that you learn, and your behaviour improves over time.”
Agarwal worked with Cenying Yang of the City University of Hong Kong and Prabhudev Konana of the University of Maryland to analyze data from a large automaker for cars sold in 2018 and 2019.
The data, collected from onboard sensors, included trips, speeds, and acceleration rates for 195,743 vehicles some with ADAS and some without.
Read the full report: “General Behavioural Impact of Smart System Warnings: A Case of Advanced Driving Assistance Systems”