• The Team

• JunHee Park - Full Stack, ParkA Database
• Jiyoung Yoon - Full Stack, Web App
• Changeun Park - Android App
• Heero Chung - Project Design and Delegation
• Luke Wulf - Hardware, IoT Solutions

The Problem

30%

Estimated cars within congested city traffic ‘cruising’ for parking.

‘Cruising’ : circling part of the city slowly in search for parking

ReinventingParking.org

20%

Estimated amount of city congestion we can eliminate with ParkA

Overview

• To build a new smart-city IoT solution
• To help locals and visitors avoid city hassles
• To improve the citywide traffic load balancing for City of San Diego
• Using state-of-the-art machine learning algorithms, smart phones, sensors with Intel IoT Kit, and measurement data with GE CityIQ

​

The Solution

• Smart-City IoT Platform
• Rapid city parking locator and suggestion
• Parking occupancy rate estimator
• Machine learning algorithm
• Supported Vector Machine (SVM)
• Android and Web integration
• Real Time ParkA Database
• GE CityIQ data (public parking) + QI IoT data (private parking)

​

Front End Back End Data Acquisition

Predix

QI Server

QI Database

< Redis: in-memory database for big data >

QI App. client

QI Web. client

Let’s Jump In and See What We Have Hacked...

Entering the App

Location Information

Filters

Scalability

Web Client

Predictions and Suggestions

• State-of-the-art machine learning algorithm
• Supported Vector Machine (SVM)

IoT

Demo

Summary

ParkA offers the city and its residents a traffic-clearing, rapid parking service.

We achieve this by:

1. Implementing City-wide traffic load balancing data
2. Using prediction algorithms to estimate future parking occupancy rates

​

Next Steps

• Launch in San Diego
• Distribute IoT modules to add more parking to map
• Expand to personal and private property owners
• Park-sharing
• More parking