Chinese car tech can spot driver’s personality traits, researchers say

Stephen Chen
·3-min read

What can your car say about you?

Using a few common sensors, BMW engineers and Chinese government researchers have developed a technology to get a general idea of a motorist’s personality from the way they drive, according to a peer-reviewed paper published in the Journal of Advanced Transportation last month.

“Participants only need to drive for less than 10km (6 miles) before their personality traits can be identified quite precisely,” the joint research team, led by Zhu Tingshao from the Chinese Academy of Sciences’ Institute of Psychology in Beijing, said in the paper.

Get the latest insights and analysis from our Global Impact newsletter on the big stories originating in China.

Zhu and BMW China recruited more than 90 volunteers in Beijing, tracking the motorists as they drove 15km through city traffic in one of the company’s cars.

The researchers compiled data from a small number of sensors that detected factors such as angle of the acceleration pedal, speed and the tilt of the steering wheel.

Each sensor measured the drivers’ movements 10 times each second. Artificial intelligence was then used to link the data to personality.

According to the researchers, the AI could predict personality in the motorists, such as whether a driver was prone to get angry or willing to take risk, with accuracy of up to 88 per cent.

Zhu’s laboratory has worked closely with the Chinese government and military to develop new technology that mines physiological traits from various forms of big data.

For instance, they found the frequency and timing of posts on social media could tell a lot about a user’s personality.

Zhu and BMW did not respond to requests for comment on Monday.

In the paper, the researchers said that knowing a driver’s personality would improve the car’s performance and safety.

“Driving signals have the advantages of being real-time, continuous, non-intrusive and reliable ... this method can monitor the continuous change of driver’s psychological indicators,” the paper said.

Including more sensors into the AI assessment would further improve the accuracy of the technology. In the future, the entire system could be built into a silicon chip and become an integrated part of the car’s on-board computer system, according to the researchers.

How artificial intelligence makes travel safer during Covid-19 and city commuting easier

Some data, such as the information stored in a car’s electronic control unit, can already be read by a mechanic. These data are usually regarded as non-sensitive and not protected by law or government regulations, according to some AI researchers.

But Zhao Yao, professor of computer science at Beijing Jiaotong University, said “some people may not feel comfortable driving a car that keeps track of their behaviour to assess what kind of person they are”.

“This is technically possible, yes, but what for, people may ask,” Zhao said.

“A car manufacturer can use the technology to improve the performance of their products, as long as they keep the data safely inside the car to prevent it from being tampered with or stolen for other purposes by unauthorised individuals or organisations.”

Last month BMW issued guidelines for the use of AI in their products, addressing a wide range of issues from safety, accountability, transparency to privacy breach.

The company would “increase awareness among its employees of the need for sensitivity when working with AI technologies”.

But carmakers did not have a perfect record on protecting their client’s privacy. The customer records of nearly 500,000 BMW, Mercedes and Hyundai owners were hacked and put up for sale on the internet in July, according to some media reports.

Technical data was not safe either. Tens of thousands factory records from Tesla, Toyota, Volkswagen and other car makers were leaked in 2018, exposing secrets in automation processes.

More from South China Morning Post:

This article Chinese car tech can spot driver’s personality traits, researchers say first appeared on South China Morning Post

For the latest news from the South China Morning Post download our mobile app. Copyright 2020.