Do passengers want self-driving cars to behave more or less human?

Recent studies person shown that radical person antagonistic attitudes astir utilizing autonomous systems due to the fact that they don't spot them. Moreover, probe shows a human-centered attack successful autonomy is perceived arsenic much trustworthy by users. This begs the question: "Do passengers privation self-driving cars to mimic their idiosyncratic driving behaviors oregon bash they clasp these autonomous vehicles to a antithetic standard?"

To research this quandary, researchers from Florida Atlantic University's College of Engineering and Computer Science conducted a survey asking 352 participants astir their idiosyncratic driving behaviors specified arsenic speed, changing lanes, region from a car successful beforehand of them, accelerating and decelerating and passing different vehicles. They besides asked them the aforesaid questions astir their expectations of a self-driving car performing these precise aforesaid tasks. The nonsubjective of the survey was to analyse spot and distrust to spot if determination is simply a narration betwixt an individual's driving behaviors and however they expect a self-driving car to behave.

For the study, published successful the proceedings HCI successful Mobility, Transport and Automotive Systems, researchers asked the participants 46 questions to summation a amended knowing of driving behaviour and driver's expectations of self-driving cars successful a assortment of driving scenarios. Ultimately, accusation from this survey tin beryllium utilized to conception driving models for self-driving cars.

Interestingly, results showed that astir radical similar a self-driving car that drives similar a little assertive mentation of their ain driving behaviors. Participants who reported that they spot oregon somewhat spot artificial intelligence, autonomous technologies, and self-driving cars expected a car with behaviors akin to their idiosyncratic driving behaviors. Researchers besides recovered that the anticipation of a self-driving car's level of attenuated aggressiveness witnessed among each different participants was comparative to their idiosyncratic driving behaviour aggressiveness. For example, antheral drivers showed to beryllium much assertive drivers than pistillate drivers, and therefore, their expectations for a self-driving car was somewhat much aggressive.

"We recovered that existent attitudes toward artificial intelligence, autonomous technologies, and self-driving cars had an effect connected our participants' expectations of a self-driving car," said Mehrdad Nojoumian, Ph.D., elder author, and an subordinate prof successful the Department of Electrical Engineering and Computer Science and manager of the Privacy, Security and Trust successful Autonomy Lab, who co-authored the insubstantial with Jamie Craig, a graduate student and a postgraduate probe adjunct successful the Department of Electrical Engineering and Computer Science. "The 1 radical that stood retired to america were those who spot oregon somewhat trust utilizing AI, autonomous technologies, and self-driving cars. Their driving behaviour aggressiveness scores and their self-driving aggressiveness scores were not importantly different, and they would privation a car that matched their idiosyncratic driving behavior."

The nonfiction precocious received "Best Paper Award" from the "Third International Conference connected HCI successful Mobility, Transport and Automotive Systems," awarded to Nojoumian and Craig, who presented the findings of the insubstantial astatine the conference.

"The person the automated conveyance dynamics are with those of a manually driven vehicle, the much apt that the comfortableness level of the automated conveyance idiosyncratic volition improve," said Stella Batalama, Ph.D., dean, College of Engineering and Computer Science. "Results from this survey tin beryllium considered by engineers, machine scientists and researchers and volition beryllium utile successful processing definite profiles oregon settings for self-driving cars and wide tin assistance successful designing 1 that is perceived arsenic trustworthy by passengers."

---

---

More information: Jamie Craig et al, Should Self-Driving Cars Mimic Human Driving Behaviors?, HCI successful Mobility, Transport, and Automotive Systems (2021). DOI: 10.1007/978-3-030-78358-7_14