Wearable AR Enhances Pedestrian Safety with AVs

Wearable Augmented Reality (AR) is a technology that could help improve how people interact with Autonomous Vehicles (AVs) when crossing streets. This technology allows users to receive real-time information right in their view, which could be useful for Pedestrians, especially in busy areas where multiple AVs are present. In this study, we examined new AR concepts aimed at helping pedestrians cross streets safely when several AVs are approaching from different directions.

#Study Purpose

The main aim of this research was to test three different AR designs that help pedestrians signal their intention to cross the street and see how AVs respond. The AR concepts were compared to a standard pedestrian push button to see which one people preferred and which one reduced the mental effort required when crossing the street.

#Importance of AV-Pedestrian Communication

For AVs to be successful, they need to communicate effectively with pedestrians. Currently, people often rely on cues like a vehicle's movement and sounds to guess what it will do. However, when a vehicle is self-driving, these cues might be unclear. To overcome this issue, researchers are looking into using external human-machine interfaces (eHMIs) that can clearly indicate a vehicle's intentions. This is particularly important for vulnerable road users, such as pedestrians.

#Scalability Challenges

Most past research focused on one pedestrian and one vehicle. However, real-world traffic often involves many vehicles and pedestrians. This makes it crucial to study how to communicate effectively with multiple AVs at once. Pedestrians may feel overwhelmed if they have to interpret signals from several AVs, which could lead to safety risks.

#Benefits of Augmented Reality

AR has the potential to improve communication between AVs and pedestrians. It can display clear information in the real world, allowing users to quickly understand their surroundings. This technology has been tested in cars to aid navigation and increase safety. Researchers are now interested in how AR can help pedestrians in complex situations, especially when multiple AVs are involved.

#Study Design

In our investigation, we created three AR concepts aimed at helping pedestrians communicate with AVs when crossing streets. The concepts varied in how they signaled responses from the vehicles. We used a virtual reality (VR) setting to test the designs against a traditional pedestrian push button.

#AR Concepts

1. Distributed Response: Each vehicle displays its own signal to indicate whether it will yield to the pedestrian.
2. Aggregated Response: A single visual signal represents all vehicles, indicating that they are aware of the pedestrian and will stop.
3. Combined Response: A mix of both distributed and aggregated signals is shown.

#Evaluation Method

The study consisted of a simulation where 24 participants crossed a virtual street. Participants tested each of the AR concepts and the push button, all while we measured their Cognitive Workload, Trust in the systems, and preferences.

#Participants

We recruited a diverse group of participants, primarily young adults familiar with technology. They experienced the VR simulation, which allowed them to walk around and notice how the visual cues worked.

#Results

#Preference for AR Concepts

Most participants preferred the combined AR concept, which used both individual vehicle responses and an overall crossing signal, over the other designs and the traditional push button. This concept reduced their mental effort during the crossing and gave them a sense of control. However, the standalone overlay concept was less favored due to lower trust and usability ratings.

#Trust in Systems

Participants generally felt more confident with the traditional push button compared to the AR concepts. Concerns about technical issues and unfamiliarity with AR impacted their trust. The combined AR concept was seen as a bridge between the traditional method and the new technology, offering more assurance in its effectiveness.

#Cognitive Load

The mental effort required for participants varied across designs. The combined AR concept resulted in lower cognitive load, making it easier for participants to cross the street without feeling rushed. Participants appreciated the clarity that the combined signals provided.

#Insights from Interviews

#Familiarity and Excitement

Participants reported feeling more comfortable with traditional methods like the push button. However, they found the AR concepts exciting and futuristic. Many felt that they would need more exposure to wearable AR before fully trusting it in real-life situations.

#Personal Device Advantages

Participants liked the flexibility that wearable AR offered compared to fixed push buttons. It allowed them to signal their intentions anywhere, but concerns about cost and data privacy were significant barriers to adoption. Some also worried about forgetting to wear or use the AR device.

#User-Initiated Communication

Everyone preferred to actively signal their crossing intentions instead of waiting for a vehicle to detect them. Participants believed that it would be safer to communicate directly, especially since they were skeptical of AVs interpreting their body language or spontaneous movements.

#Technical Concerns

Many participants raised concerns about the technology's reliability. They were wary of potential failures, such as poor communication between the AR glasses and the AVs. Some suggested that a system of shared visual signals among users could help improve overall safety.

#Conclusion

The study found that wearable AR technology has great potential for improving pedestrian safety at street crossings involving AVs. The combined AR concept performed best, reducing cognitive load and increasing trust. However, barriers such as cost and data privacy need to be addressed to promote wider adoption. Future research should consider how to enhance communication strategies to better serve both pedestrians and AVs while ensuring safety.

#Future Directions

Future studies should aim to involve a broader and more diverse group of participants. The technology used in the VR simulations may also be refined to better replicate real-world conditions. Exploring how AR can be effectively integrated with existing infrastructure to improve AV-pedestrian interaction will be essential for the technology's success. Additionally, understanding how different cultures perceive and interact with AR systems will be crucial as this technology evolves.