This 'staring car' with robotic eyes recognizes pedestrian signals - here's how

This ‘staring car’ with robotic eyes recognizes pedestrian signals – here’s how

“Self-driving cars are the future, but they are based on predictions, algorithms, or sensors. The AI ​​of the car system doesn’t interact much with pedestrians. And that’s a problem. That was our starting point,” Igarashi tells us. Interesting geometry (IE) in an interview.

The researchers felt that there had not been much research or investigation into the “human” component of self-driving technology, even though the concept had been moderately explored before.

Robotic eyes were installed on a golf cart for the experiment.

Enabling effective interaction between pedestrians and vehicles

“We’ve noticed that people who drive are using gestures, or are communicating in some way with pedestrians who might be on the road. We wanted to do something similar for self-driving cars. Then again, the communication possibilities, like voice or movement, abound,” says Igarashi. .

Then the researchers considered another factor. The driver becomes more than a passenger in self-driving vehicles, which is a “key difference”. As a result, they either pay little or no attention to the road and look away, or there is no one behind the wheel physically.

This makes it difficult for pedestrians to determine if a self-driving car has registered their presence, as there may be no eye contact or contact from people inside.

So, how do pedestrians know if a car has noticed their presence? Igarashi explains: “We decided to use the eyes or squinting as a means of communication between vehicles and people. Our focus was on ‘attention.’”

As an experiment, a self-driving golf cart was equipped with two large, remote-controlled robot eyes, aptly nicknamed the “staring car”. The team wanted to test whether placing moving eyes on a vehicle would affect pedestrian behavior and whether people would continue to cross in front of a moving vehicle when they were in a hurry.

you know this

The staring car notices a pedestrian in a virtual reality scenario.

However, there was a catch. It would be very dangerous to ask volunteers to choose whether or not to walk in front of a moving vehicle in real life.

Igarashi continues, “Currently, the eyes are mostly shown. In this evaluation, the eye movement was manually controlled by someone inside the car.”

Behind the scenes with buggy and googly eyes

Participants played four scenarios, two with the carriage with eyes and two without, using virtual reality. Then they had to decide whether or not to cross a road in front of a moving vehicle.

When the virtual car was equipped with robotic eyes, which either looked at pedestrians (registering their presence) or away (not registering them), participants were able to make safer and more effective choices. The wagon either noticed the pedestrians and was thinking of stopping or did not notice them and would continue to drive. His eyes are either looking toward or away from the pedestrians.

you know this

n (a) the cart takes care of the subscriber (it can be safely crossed); in (b) the cart does not interest the subscriber (unsafe to cross); In (c) and (d) the participant is not known.

Scenarios were recorded using 360-degree video cameras and 18 participants (nine women and nine men, ages 18 to 49, all Japanese) participated in the experiment. They were given three seconds each time decide whether Or they don’t choose to cross the road in front of the wagon.

The researchers recorded their choices and measured error rates in their decisions — how often they chose to stop when they could cross, and how often they crossed when they had to wait.

They noticed a clear difference between the sexes, which they found “surprising.” Other factors such as age and background may also have played into the participants’ reactions and the results show that road users are subject to different behaviors and needs that require different ways of communication in the world of self-driving.

Male participants made many dangerous decisions to cross the road, such as choosing to cross when the car would not stop, but the eye of the cart reduced these errors.

“Participants did not make dangerous choices, but were so careful that they stopped even when the car stopped. At the same time, they said that they felt safe when the eyes of the cars were on them. In general. The results were very interesting,” says Igarashi.

Experience proved that the eyes led to a safer crossing for all.

Will we have cars that can “see” us?

Experience has its limits, Igarashi says. For starters, it was performed in a VR environment, which makes it difficult to accurately measure the data.

“Real life is a lot different. Arguably our group was also small. Most importantly, all of our participants were Japanese. The Japanese tend to be very careful when it comes to road safety. The scenario would be very different in other countries. But I think our research was A good starting point,” says Igarashi.

“People really liked this kind of thing – some said the car was cute, the rest said it looked creepy. Regardless, I could see that the contribution to safety was generally appreciated,” continues Igarashi while discussing responses to his research.

Next, the researchers intend to work on developing automated control of robotic eyes connected to self-driving AI (instead of manually controlled), which can accommodate different situations.

“If eyes can truly contribute to safety and reduce traffic accidents, then we must seriously consider add them. But it is very difficult to predict when it will become a reality. As someone in academia, we’re trying to work on what could become necessary and realistic in 10 years or so, when self-driving technology matures.”

you know this

A 3D image of a modern car powered by artificial intelligence to assess driving conditions and street elements.

Over the past few years, makers of self-driving cars have raised a lot of money based on promises to develop a fully automated product. But recently, industry leaders and experts have said that technology will always be Requires human supervision.

“This is a difficult question. In short, producing a fully autonomous car is very difficult. Human support is necessary for most public situations,” argues Igarashi.

He continues: “Regardless, we should think more about the interaction between pedestrians and vehicles. That was the most important part of my research. It’s futuristic, but this kind of communication technology is absolutely essential,” he adds.

Study summary:

Several car manufacturers and researchers have explored the idea of ​​adding eyes to the car as an additional method of communication. A previous study showed that the eyes of self-driving vehicles help pedestrians make decisions to cross streets faster. In this study, we examine a more important question, “Can eyes reduce traffic accidents?” To answer this question, we consider a critical street crossing situation in which a pedestrian is in a hurry to cross the street. If the car is not looking at pedestrians, this means that the car does not recognize pedestrians. Thus, pedestrians can be judged not to cross the street, thus avoiding potential traffic accidents. We conducted an experimental study using 360-degree video imaging of a real car with robotic eyes. The results showed that the eye can reduce potential traffic accidents and the direction of gaze can increase the pedestrian’s personal feelings of safety and danger. In addition, the results showed gender differences in critical and non-critical scenarios in AV-to-pedestrian interaction.

#staring #car #robotic #eyes #recognizes #pedestrian #signals #heres

Leave a Comment

Your email address will not be published. Required fields are marked *