graduate school · Master's thesis


Enhancing Traffic Safety by Preventing Driver Anger and Saving 3,300 Lives Annually with Loved One's Voice Messages

graduate school · Master's thesis


Enhancing Traffic Safety by Preventing Driver Anger and Saving 3,300 Lives Annually with Loved One's Voice Messages


- Calmdrive, my MFA UX Design thesis project at the College for Creative Studies, aims to prevent road rage and enhance driver safety. Witnessing and experiencing road rage in the US inspired me to create this project. Calmdrive detects physiological signs of anger in drivers and calms them with a pre-recorded message from a loved one. The data is sent to the Calmdrive app, where AI offers recommendations to prevent future anger.


- To design a system that detects and prevents anger while driving, thereby enhancing road safety.


- Programming sensors using Arduino demanded a significant amount of time and posed numerous challenges for me.


- Demonstrated the feasibility and effectiveness of using loved ones' voice messages to reduce driver anger by an average of 25% in heart rate and 27% in stress level.

- Validated the potential for significant real-world impact in enhancing traffic safety and reducing fatalities caused by road rage.


- The potential exists not only for the reduction of road rage-related accidents but also for the saving of 3,300 lives annually in the United States.

- The potential to help mitigate increases in insurance premiums due to road rage.


- 100% individual project: from
research, user interviews,
and usability testing to prototypes.


- Figma
- Arduino
- 3D printing


- Mobile and IOT


- September 2023 ~April 2024

Final Prototype Video


To prevent angry driving, Calmdrive offers two solutions

1. Sensors and Voice Message

Sensors in seatbelt and wristband detects driver's anger signal and plays a voice message from a loved one.

2. 3 Key Features of the Calmdrive Mobile App

After driving, drivers can track physiological data, receive personalized recommendations, and customize the voice message through the app.

Tracking physiological data
Personalized recommendation
Recording & customizing voice message


Road rage shootings in the United States have increased by 449% over the past decade.

According to research, driving while angry leads to various risks, such as poor decision-making, decreased driving performance, slowed reaction time, and loss of self-control. To make matters worse, Between 2014 and 2023, the number of people shot in road rage incidents surged more than 400 percent, from 92 to 481.


When drivers become angry on the road due to traffic situations or rude behavior from other drivers, the likelihood of serious accidents, including injuries or fatalities, increases. Preventing drivers from falling into anger problems could lead to safer traffic.


I interviewed 9 US drivers to better understand the problem and discovered key findings and insights that informed my design decisions.

1. Preventing anger before the driver enters the anger zone

Because, No driver wants to drive angry, as it can lead to unpredictable consequences.

2. Connecting with emotionally close individuals

Because, Passenger intervention, especially from family members, can help drivers calm their anger.

3. Diverting attention from anger to focus on driving

Because, Anger makes it hard for drivers to focus, as they keep thinking about why other drivers act a certain way.

Driver Needing Anger Prevention
A driver living in Michigan who commutes daily to support his family.

- Sometimes, he gets angry due to traffic situations or other drivers' rude behavior.

- He wants to prevent anger before it escalates into road rage.

- The intervention of a loved one helps him refocus on driving when he is angry.


The solutions currently available in the market to help manage driver anger are limited.

The academic and automotive industries have proposed various solutions, such as music playback based on real-time emotion recognition, blue light, and voice assistants. However, drivers who are already angry may not respond well to these solutions, and complex technologies like emotion recognition are still very expensive to implement in cars.

01. KIA’s Emotion Adaptive Driving
02. Valeo’s Interior Lighting
03. Visage SDK’s Face Tracking


I developed two prototypes that are more affordable and effective in preventing anger.

Current solutions are either ineffective for angry drivers or too costly to adopt widely. To address this, I developed prototypes to prevent road rage before drivers reach the anger zone, all within a $100 budget.

1. Gadgets: Detects anger sign while driving and plays a loved one's voice message.

The seatbelt pulse sensor measures heart rate, and the wristband GSR sensor gauges stress levels through skin conductivity. When anger starts, heart rate and stress levels rise.

01. Seatbelt, Pulse sensor, Speaker(voice message)
02. Wristband, Galvanic skin response sensor, Display

The process of creating gadget prototypes

I programmed with Arduino and created the case for the Arduino and display using a 3D printer.
After more than 10 failures and over 200 hours of effort, I was finally able to create a prototype to test.

One big challenge

Using pulse rate and GSR (Galvanic Skin Response) sensors to accurately detect a driver's anger state was a significant challenge. I especially focused on ensuring the reliability and accuracy of sensor detection in real driving situations without distracting the driver. To achieve this, I conducted repetitive tests, adjusting the sensor positions and modifying the code, in order to find the optimal settings.

2. Mobile app with personalized solutions for preventing anger before and after driving.

Data collected from the gadget is transmitted to the mobile app for integrated management. This allows drivers to track their physiological data, receive personalized AI recommendations, and record and customize voice messages to prevent angry driving.

01. Tracking physiological data

Tracking driver's physiological data while driving can help them become more aware and manage their emotions better.

01. Sensors detecting
02. Dashboard
03. Heart rate tracking
04. Stress level tracking

Improving the dashboard

- I introduced a character to intuitively show the driver's mood, but the primary target users of Calmdrive are adult drivers. Therefore, the cute character style could feel childish to them, reducing the credibility of the information.
- To address this issue, I replaced the character with colors and numbers to indicate how calm or angry the driver is. As a result, users reported feeling more credibility in the information compared to the previous version.

02. Personalized Recommendation

CalmDrive Assistant's personalized recommendations provide drivers with actionable recommendations based on physiological data collected by sensors during driving.

01. Recommendations from calmdrive assistant
02. Recommendations from heart rate

Improving personalized recommendations from heart rate

- Blue for recommendation maintained brand identity but made it seem less important since blue often signifies positive information.
- The prompt 'Tap here to view recommendations for you' leads to unnecessary clicks, failed to convey important information immediately.
- Personalized recommendation carries warnings for drivers. Therefore, I changed the color from blue to red, as red better conveys warnings.
- The prompt ‘It's crucial to stay calm after 9:15pm’ increased user satisfaction by offering specific, personalized recommendations, making the app feel more tailored.

03. Recording & Customizing Voice Message

Drivers can customize mobile app to prevent anger by recording the voice of a loved one, using existing samples, or selecting AI prompts.

01. Voice message dashboard
02. Selecting options
03. Using samples
04. Generating prompts

Improving the voice message dashboard

- Drivers had to keep seeing recording tips even after checking.
- Recording tips and action buttons are necessary but not frequently used features.
- Recording tips, recorded messages, and action buttons made prioritization difficult for drivers.
- I moved the recording tips to a top banner, allowing drivers to view them only when they want.
- Since drivers frequently use recorded messages, I made it the main function for easy access, removing the need to click "more."
- I replaced the two action buttons with a single + action button that reveals more options when clicked.

A voice message from a loved one reduced heart rate by an average of 25% and stress level by 27%.

I selected drivers in the US to wear the gadgets while watching a road rage video to induce anger. During this time, a pre-recorded voice message from a loved one was played, and their physiological signs of anger decreased.

Final usability testing
Stress level(µS)
Pulse rate(BPM)
IMpACT to the real world

If the Calmdrive system were implemented to prevent road rage incidents,

1) It could potentially save approximately 3,300 lives annually.

According to the NHTSA report, 40,990 people died in traffic accidents in the U.S. in 2023. With 66% of fatal traffic accidents related to road rage, if 25% of the 220 million drivers use the Calmdrive system and it has a 50% success rate, overall traffic accidents could decrease by 8.25%, potentially saving around 3,

2) It could help mitigate increases in insurance premiums.

Aggressive driving is not considered an accident but a dangerous driving behavior. Therefore, auto insurance companies do not cover road rage-related incidents as losses or liabilities. In Hawaii, auto insurance premiums can increase by up to 96% after receiving a fine for road rage-related behavior, with similar increases seen in California (74%), Michigan (69%), North Carolina (65%), and New Jersey (54%). Implementing systems like Calmdrive that prevent road rage and improve driver safety can address these issues.

thesis defnse

Thesis Defense Success: Calmdrive Validated

I successfully defended my thesis before a panel of internal and external experts, who acknowledged the feasibility and usefulness of the Calmdrive system.

Thank you for scrolling!

What I've learned

I explored various dimensions of UX design, from technical components to design principles.

1. This project was conducted independently, where I undertook the entire UX design process from start to finish. This included research, problem definition, conducting interviews, prototyping, and usability testing. Managing each step on my own has proven my capability to execute a project independently.

2. Creating a physical prototype using Arduino, a pulse sensor, and a GSR sensor allowed me to test the system's functionality in real-world scenarios. This hands-on approach provided valuable insights that could not have been gained through digital simulations alone.

3. The process of iterating on the design and conducting usability tests was crucial. Each round of testing uncovered new issues and provided actionable feedback, leading to significant improvements in the final product.

More in the Desktop.

You are now reading the mobile version of the site. For detail case study, visit with a desktop browser :)

About the project
Competitive analysis