A Social Approach for Autonomous Vehicles: A Robotic Object to Enhance Passengers’ Sense of Safety and Trust

One of the central challenges in designing autonomous vehicles concerns passenger trust and sense of safety. This challenge is related to passengers' well-established past experience with non-autonomous vehicles, which leads to concern about the absence of a driver. We explored whether it is possible to address this challenge by designing an interaction with a simple robotic object positioned on the vehicle's dashboard. We leveraged the automatic human tendency to interpret non-verbal robotic gestures as social cues and designed an interaction with the robot in the autonomous vehicle. The robotic object greeted the passenger, indicated that the vehicle was attentive to its surroundings, and informed the passenger that the drive was about to begin. We evaluated whether the robot's non-verbal behavior would provide the signals and social experience required to support passengers' trust and sense of safety. In an in-person (in-situ) experiment, participants were asked to enter an autonomous vehicle and take the time to decide if they were willing to go for a drive. As they entered the vehicle, the robot performed the designed behaviors. We evaluated the participants' considerations and experience while they made their decision. Our findings indicated that participants' trust ratings and safety-related experience were higher than those of a baseline group who did not experience the interaction with the robot. Participants also perceived the robot as providing companionship during a lonely experience. We suggest that robotic objects are a promising technology for enhancing passengers' experience in autonomous vehicles.