Autonomous cars equipped with giant eyes would reduce the risk of accidents with pedestrians

Autonomous cars equipped with giant eyes would reduce the risk of accidents with pedestrians

⇧ [VIDÉO] You might also like this affiliate content (after advertising)

Each year, around 15% of people killed on French roads are pedestrians. Mutual visibility and anticipation issues between pedestrians and drivers are largely responsible, raising real safety questions in the age of driverless autonomous cars. Recently, a team of Japanese researchers developed a concept for a “surveillance” car with robotic eyes that allows a pedestrian to understand the car’s behavior. This innovation could equip future autonomous cars, the design remains to be defined.

The development of autonomous vehicles is rapidly bringing us into a new era of transportation. Whether it’s delivering packages, plowing fields, or transporting people within a campus, there’s a lot of research going on to put this concept into practice while keeping road traffic safe.

But one of the main differences between conventional and autonomous vehicles is that the drivers of the latter can become more passengers, so that they no longer pay full attention to the road, or that there is none – just being nobody behind the wheel. This makes it difficult for pedestrians to judge whether the vehicle has detected their presence or not, as no eye contact or indication of people inside may be visible.

That is why scientists from the University of Tokyo studied more precisely this concern affecting the very relationship to the autonomous car, which should be more “human”. They wanted to find a way to alert pedestrians when an autonomous vehicle spots them and intends to stop, or when it hasn’t seen them and therefore doesn’t stop when they pass. A paper on their research was presented at the 14th International Conference on Automotive User Interfaces and Interactive Vehicle Applications.

A car straight out of a cartoon

To overcome the problem of “coupling” between the car and the pedestrian, the researchers equipped the autonomous golf cart with two large robotic eyes controlled manually by the driver, like a cartoon car. Thanks to the one-way mirror film on the windshield, the interior of the vehicle was hidden from the view of pedestrians, giving the impression that there was no driver. Scientists called it ” Staring car (watch the car). They wanted to test whether sticking rolling eyes on a vehicle would affect people’s risk-taking behavior, i.e. whether people always cross the road in front of a moving vehicle when they are in a hurry.

Takeo Igarashi of the Graduate School of Information Science and Technology reports va communicated : ” There is not enough research into the interaction between self-driving cars and the people around them, such as pedestrians. Therefore, we need more research and effort on such interaction to provide safety and reassurance to society when it comes to self-driving cars. “.

The team prepared four scenarios, two where the vehicle had eyes and two where it did not. Either the car noticed the pedestrian and intended to stop, or it didn’t and was going to continue driving. When equipped with eyes, they either looked towards the pedestrian to signify ” i will stop » or looked elsewhere, which means « i won’t stop “.

In the experiment, participants have to decide whether the vehicle has noticed them and will stop. The images show the participant’s first-person view. In (a), the vehicle pays attention to a participant that is safe to cross; vb) does not pay attention to a participant who considers the transition dangerous; and in letters c) and d) the participant does not know. © Chang et al. 2022.

Eighteen volunteers—nine women and nine men aged 18 to 49, all Japanese—conveniently participated in the study. The authors decided to test their attitude thanks to augmented reality, so as not to take risks in a real situation.

This is how the team recorded the scenarios using 360-degree video cameras. Using augmented reality, volunteers safely experienced the scenarios 40 times in random order, each time having three seconds to decide whether or not to cross the road in front of the vehicle. The researchers recorded their choices and measured the error rate, that is, how often they chose to stop when they could cross, and how often they crossed when they should have waited.

virtual reality car test
A video of a virtual reality experience in which participants cross a road on the campus of the University of Tokyo. © Chang et al. 2022.

Different results between men and women

After analyzing the results, a clear difference in responses between the sexes emerges. Chia-Ming Chang, project lecturer and member of the research team explains: Although other factors such as age and background also influenced participants’ responses, we believe this is an important point because it shows that different road users may have different behaviors and needs that require different ways of communicating in our future world in which we manage ourselves. “.

Specifically, males made the greatest number of unsafe decisions to cross the road (ie, decided to cross when the car did not stop), but these errors were reduced by looking at the vehicle. On the women’s side, the authors point out that they made ineffective decisions, in other words, they chose not to pass when the car intended to stop. But these errors were also reduced by the appearance of the vehicle.

However, the presence of eyes did not significantly change the responses of men in safe situations (i.e., choosing to cross when the car was about to stop) or women in dangerous situations. They didn’t really go over whether the vehicle had eyes or not. The authors conclude that the eyes result in a smoother or safer crossing for all.

Humanized car reaction

The authors then looked at what the eyes evoked in the participants to explain the different results obtained. Feelings are just as mixed as the previously mentioned reactions.

That’s how some volunteers found them (eyes) cute while others saw them as scary. For many male participants, when their eyes were averted, they reported feeling that the situation was more dangerous. For women, when eyes were on them, many said they felt safer.

Igarashi says: ” We focused on eye movements, but in this particular study we did not pay too much attention to their visual design. Due to budget constraints, we kept it simple to minimize design and construction costs. In the future, it would be better for a professional product designer to find the best design, but it would probably be difficult to satisfy everyone anyway. “.

The team acknowledges that this study is limited by the small number of participants playing a single scenario. It is also possible that people make different decisions in virtual reality than in real life. However, the era change that is looming in the future with these autonomous cars requires additional safety points, and the realization of such studies will make driving on the road safer for all users, whether they are motorized or not.

Researchers note: In the future, we would like to develop autonomous automatic control of the AI-connected robotic eye (instead of manual control) that could adapt to different situations. “.

Source: AutomotiveUI ’22: Proceedings of the 14th International Conference on Automotive User Interfaces and Interactive Automotive Applications

Author Image

Leave a Reply

Your email address will not be published.