The Impact of Vehicle Appearance and Vehicle Behaviour on Pedestrian Interaction with Autonomous Vehicles

What and why?
1. Existing research suggests that vehicle appearance plays a significant role in the perception of the vehicle’s sociability and power1 .
2. Vehicle design can affect the car’s perceived aggression or assertiveness, or on contrary, its cuteness or friendliness. The looks of a vehicle often lend themselves to stereotypes about the kind of people who own them, and their driving behavior2,3.
3. A pedestrian may look at a vehicle and instantly make assumptions about the vehicle’s assertiveness, the potential to yield, etc., and consequently determine how to interact with it.

With autonomous cars: How will the appearance of the vehicle will impact the perception of it, and subsequent interaction with it?

Auto manufacturers employ numerous design techniques to help their concept vehicles stand out.

1. Faraday Future calls out its vehicle FF91’s design and draws attention to its futuristic features.
2. The Mercedes Benz F015 Concept vehicle
3. BMW Vision Next 100 Concept
4. Nissan IDS Concept
5. Ideo’s Future of Automobility Concept
6. Volkswagen’s Sedric Self-Driving Concept

Car manufacturers highlight their autonomous vehicles to be futuristic and capable of much more than what people would expect of an ordinary car. Will there be a side effect to such remarkably futuristic designs?

If a vehicle is autonomous, irrespective of how the vehicle looks from the outside, the behavior of the vehicle is controlled by the technology and not by a human driver. Then, will the vehicle’s appearance still have a direct correlation with the stereotype of the kind of driving behavior expected from it?

What message does a vehicle’s appearance convey to its surrounding? While auto manufacturers are eager to set themselves apart as pioneers of future technology and lifestyle solutions, it remains to be seen if designs that are remarkably different from what people are used to seeing around today will affect people’s mental models of how to react in the presence of such vehicles and preserve safety.

Contribution of this research
We attempt to answer a fundamental question of the impact of vehicle design within the context of vehicle autonomy. The results can play a role in identifying the factors responsible for pedestrian interaction with vehicles, and in suggesting the direction of the design and communication paradigm that autonomous vehicles should adopt to ensure a non-disruptive acceptance into society.

Study
Two-part analysis Survey
A questionnaire showing 10 different cars in front view with a pedestrian next to it for scale. The questionnaire asked the participants to rate the cars on their appearance, based on their perception of friendliness vs. aggressiveness, ordinariness vs. futuristic-ness, etc. The Renault Twizy was chosen to be a representative of a friendly-faced, futuristic-looking vehicle. The BMW 3 series was chosen as an aggressive-faced, ordinary-looking vehicle after collecting responses from 100 participants.

Experiment
The two cars chosen were used for the subsequent experiment.
1. The vehicles were driven on a straight, otherwise empty road
2. They were filmed from the point-of-view of a pedestrian standing on the edge of the road (picture on the left)
3. The road did not have a pedestrian crossing
4. At various pre-determined distances from the pedestrian, the video was clipped. Each short video segment of the approaching car was shown in a random order to participants, and they were asked to indicate their willingness to cross the road on a 5-point Likert scale

Independent variables:
1. Vehicle appearance (the two vehicles)
2. Vehicle behavior (assertive vs. yielding behavior)
3. Vehicle driving mode (manual vs. autonomous driving)

Dependent variable:
The willingness of pedestrian to cross the road

Study design:
1. Between-subjects design among manual and autonomous driving conditions
2. In manual driving condition, the driver wore a high visibility jacket to make his presence obvious
3. Autonomous driving condition was simulated using the Ghost Driver Wizard of Oz setup4 (Seat suit – picture below).

The study was concluded with an “Assertiveness questionnaire”, and a post-experiment interview about their experience. This qualitative data informed our preliminary results

References
Sonja Windhager, Dennis E. Slice, Katrin Schaefer, Elisabeth Oberzaucher, Truls Thorstensen, and Karl Grammer. 2008. Face to Face: The Perception of Automotive Designs. Hum. Nat. 19, 4 (2008), 331–346.

Graham M. Davies. 2009. Estimating the speed of vehicles: the influence of stereotypes. Psychol. Crime Law 15, 4 (2009), 293–312. DOI:[ Ссылка ]

свернуть