WTA Traffic Mobile App

Get Started with App

WTA Mobile Application empowers users in North America to report road conditions accurately and efficiently using cutting-edge AI-powered road detection and traffic element recognition technology.

Project Vision

This project aims to revolutionize traffic safety by integrating a machine learning-powered traffic condition detection framework into the WTA mobile application.

- Build an iOS detection framework that empowers developers with a flexible, scalable solution to create traffic-related applications.
- Develop a cross-platform Road Condition Flutter App with machine learning capabilities, enabling real-time detection of road conditions for both iOS and Android platforms.

My Role

Lead iOS Engineer

UX Engineer

Timeline

2024 Q2 - Q3

Project Strategy

ENABLE transportation departments, WHO lack the resources for timely and accurate road condition reporting, TO efficiently detect and manage reports BY integrating a machine-learning-based road condition detection system for automated, real-time insights.

My Contributions

During the WTA mobile app upgrade project, I led and executed 2 pivotal feature implementations:

iOS ML Road Condition Framework:
- Designed and developed a modular framework in Swift to enable real-time road condition detection.
- Integrated machine learning models using CoreML, achieving 91% accuracy in road condition predictions.
- Ensured scalability and reusability by creating a plugin-based architecture for cross-platform compatibility.
Road Condition Flutter App:
- Built a cross-platform app in Flutter to bring real-time road condition detection to both iOS and Android devices.
- Work as UX designer to create intuitive user flows and interfaces that simplify ML-based detection and reporting.
- Reduced development time by 50% through a unified codebase while maintaining native performance.

Brainstorming & Research

How did this project start?

How to solve?

- 🧠 Innovation

How can we enable real-time road condition reporting in challenging environments using machine learning?

- 💰 Cost Control

How can we reduce the cost / effort to maintain the codebase since there're lots of new project coming soon and we have limit sources in our team?

- 📶 Scalability

How can we expand our product line in a scalable and maintainable way?

Part 1

iOS Road Detection Framework

A library enables real-time road condition detection and analysis. It is designed to be integrated into various platforms, providing developers with the flexibility to build traffic-related applications.

How did we build it?

Part 1

iOS Road Detection Framework

What does it include?

3 Core Functionalities

I played a pivotal role in designing and implementing the WTA Mobile Application's core features. These advanced functionalities were developed to enhance road condition reporting and traffic management through cutting-edge technology

🚙 Road Condition Predictor:
- Self-trained Classification model
- Support video streaming detection
- Built on Tensorflow Framework
🚦 Traffic Element Recognizer:
- Real-time Object Detection model
- Support video streaming detection
- Built on Tensorflow Framework
📱OBDII Scanner:
- Bluetooth Connection
- Support ELM327

Self-generated Framework Documentation

RCDetector Framework Documentation, which is natively generated directly from the framework’s codebase. By leveraging automated documentation tools, it provides an up-to-date and comprehensive resource for developers to seamlessly implement and integrate the framework into their projects.

Part 2

Road Condition Flutter App

🛣 Highway Reporter Functionalities:
- Includes core features for our current traffic condition reporting application, such as mode switching, detailed data views, and customizable settings.

🔁 Machine Learning Integration:
- Leverages the RCDetector Framework to enable AI-driven road condition predictions and traffic element recognition in real-time.
🛠 Cross-Platform Compatibility:
- Developed with a single Flutter codebase, ensuring consistent performance and reduced development effort for multi-platform deployment.

Our new app fulfill all functionalities from current product

Provide ML-based option for auto road condition reporting

Real-time Demo

This demo highlights the end-to-end workflow for generating road condition reports using two distinct approaches:
- Manual Reporting from Record
- New ML-Based Auto Reporting

‍In the demo, we simulate a driving scenario using a winter road video stream. The new ML-based method leverages a camera to stream real-time road conditions at 2-second intervals, enabling automated data processing through an embedded machine learning model. The system processes this real-time data and returns accurate road condition results seamlessly.

How did we build it?

Solutions

Utilizing the RCDetector Framework

Imported the RCDetector Framework as the backbone for road condition detection and ML-based features.

Step 1. Import the Framework

Integrated the RCDetector Framework into the app’s structure to enable modular functionality.

Step 2. Build the Bridge

Established connections between the Flutter app and the framework using native code bridges.

Step 3. Call Framework Functions

Implemented key functionalities like prediction models and object detection into the app's workflow.

What's on going?

The OBDII Scanner functionality is fully operational and integrated into our iOS framework with ELM327 chip support. The next step is to incorporate it into our Flutter app, enabling the collection of additional vehicle data to enhance the accuracy of road condition reporting.

[data-wf-bgvideo-fallback-img] { display: none; } @media (prefers-reduced-motion: reduce) { [data-wf-bgvideo-fallback-img] { position: absolute; z-index: -100; display: inline-block; height: 100%; width: 100%; object-fit: cover; } }

🔮 Product Vision -> ✨ My Value

Strategy <-> Tech <-> Collaboration

1. Strategic Vision 💡

- Designed a framework with scalability in mind, broadening its user base to support developers across industries.
- Incorporated multi-platform capabilities, making the solution accessible to both iOS and Android ecosystems.

2. Technical Expertise 💻

- Took on multiple roles as a UX Designer and iOS Developer. Independently developed a Flutter app from 0-to-1, seamlessly integrating it with a high-performance iOS framework to ensure scalability and flexibility.

3. Problem-Solving 🧠

- Anticipated and addressed challenges related to multi-platform compatibility and real-time data processing, delivering a robust and efficient framework.

4. Collaborative Development 🤝

- Delivered documentation and resources (natively generated) to empower a broader developer community, enhancing adoption and usability.

WTA Traffic Mobile App

Project Vision

My Role

Timeline

Project Strategy

My Contributions

During the WTA mobile app upgrade project, I led and executed 2 pivotal feature implementations:

How did this project start?

How to solve?

iOS Road Detection Framework

A library enables real-time road condition detection and analysis. It is designed to be integrated into various platforms, providing developers with the flexibility to build traffic-related applications.

How did we build it?

iOS Road Detection Framework

What does it include?

3 Core Functionalities

Self-generated Framework Documentation

Road Condition Flutter App

Our new app fulfill all functionalities from current product

Provide ML-based option for auto road condition reporting

Real-time Demo

How did we build it?

Utilizing the RCDetector Framework

Imported the RCDetector Framework as the backbone for road condition detection and ML-based features.

Step 1. Import the Framework

Step 2. Build the Bridge

Step 3. Call Framework Functions

What's on going?

🔮 Product Vision -> ✨ My Value

1. Strategic Vision 💡

2. Technical Expertise 💻

3. Problem-Solving 🧠

4. Collaborative Development 🤝

More projects