Abstract

Objective: This study aimed to improve overall traffic flow and enhance the safety of both personal mobility users and other road users by developing a smart helmet capable of real-time interaction with the driving environment.

Background: The rapid increase in personal mobility users has led to a corresponding rise in related accidents, driving the growth of the smart helmet market. However, current smart helmets predominantly offer convenience features such as Bluetooth speakers and microphones, with few providing safety functions to actively prevent accidents. To address this gap, this research focused on developing a smart helmet with real-time interaction capabilities designed to prevent accidents proactively.

Method: Following user experience research and analysis, three primary user needs were identified:

1. The ability to signal left or right turns while riding.

2. Awareness of vehicles approaching from behind.

3. The ability to indicate stopping or deceleration intentions.

The primary focus was on signaling turns. Personal mobility devices, particularly electric scooters, have low rotational inertia, making it risky for users to remove a hand from the handlebars to operate hand or foot buttons. As such, the study hypothesized that head gestures, a relatively free form of input while riding, would be superior to other methods for signaling turns. A within-subject experiment was conducted with 20 participants in their 20s, comparing three input methods: head gestures, hand buttons, and foot buttons. These were evaluated across three criteria: safety, naturalness, and satisfaction, using a 5-point Likert scale and open-ended questions. Additionally, a prototype helmet incorporating turn signals, rear approach warning lights, and emergency braking lights was developed for testing.

Results: In the turn-signaling experiment, 12 out of 20 participants rated head gestures as the most natural, and 17 out of 20 rated them as the safest. Analysis of variance (ANOVA) for naturalness showed a significant p-value of 0.013. Post hoc comparisons indicated that head gestures were not inferior to hand buttons, whereas foot buttons were significantly less natural. The results showed a significant difference between head gestures and foot buttons (p = 0.31) but no significant difference between head gestures and hand buttons (p = 0.992). For safety, ANOVA results indicated a significant p-value of 0.001. Post hoc analysis revealed that head gestures were safer than other methods, while foot buttons were less safe (p = 0.002). Satisfaction scores showed no significant differences among methods (p = 0.062).

Conclusion: The smart helmet allows personal mobility users to safely operate turn signals using the innovative input method of head gestures. Users can communicate their driving intentions with turn signals and emergency braking lights while receiving information about their surroundings through rear approach warning lights. This real-time interaction reduces the likelihood of accidents on the road.

Application: By providing real-time interaction and environmental feedback, the smart helmet enhances both physical and psychological safety for users. Additionally, other road users can better anticipate and respond to the helmet user’s driving intentions, improving the overall traffic experience.