Business Plan

Quadcopter Parking Enforcement

Authors (QuadSquad): Alexander Sergian, Kelvin Lu, Mark Machado, Baotuan Nguyen

Introduction

Goals and Deliverables: The goal of this project is to repeatedly fly our drone over a designated parking area and monitor parking times. We want to enforce against over-parking in spots designated to be parked at for only a few hours at a time.

Significance: For most parking spots with time-constraints, you rely on a person to drive around parking lots and monitor how long people have been at certain parking spots. With the drone, this individual can simply sit and watch a live feed. The drone will monitor license plates, and when a certain license plate has over-parked, it will display on the headset the license plate and location so the officer is alerted. The info will also go into a server in case the officer is busy at a certain time. He can then refer to this database to access current over-parked car information. He saves scanning time and ensured correct ticketing. This greatly enhances parking enforcement. Better enforcement would also reduce parking abuse, making it a better experience for everyone.

Background: As far as we have found, no one has addressed this exact problem.  There are groups who have addressed similar problems though.  Dr. Hui Feng and a team from Fudan University used a DJI Matrice 100 drone with some additional hardware to detect illegally parked cars (parked over the lines).  A Houston based startup called ParkiT has software that uses parking lot cameras to detect what spots are filled in a lot using computer vision and machine learning techniques.

System architecture overview

We will have the following technologies: DJI Phantom 3 quadcopter, headset display, a camera on the quadcopter, and a server. The quadcopter will scan the images, send them over a WiFi connection to a phone. We will develop a mobile app to monitor that connect and perform object recognition and subsequent extraction of license plate information. It will then send data to a server via a text API; the server will parse that into MySQL and perform associated database transactions to keep tabs of the system. The server will then keep a tab on how long each license plate has been parked and where the vehicle was parked (GPS coordinates); it will also remove license plate information once that plate hasn’t been found in a round (meaning the car left). Once the server finds an over-parked car, it will store it in a separate database and send over a notification to the headset display of a city parking officer with the current time and the location and image of the over-parked car. The information will also be sent to the officer’s mobile app in the case that he doesn’t have his headset display with him. Moreover, we will pre-emptively text a reminder to a user to request that they move their car before they get cited; the reminder will be sent at a time set by the user.

Winter Quarter bare-bones prototype Goals:

  • Manual flying of quadcopter; captures should be sent to a mobile app on a phone connected to the DJI controller. (February 22, 2016)
  • Phone app will perform OCR recognition on picture (no segmentation / locating of license plate until waypoint functionality is setup in Spring Quarter). (February 29, 2016)
  • Text API communication between mobile app and server. (February 29, 2016)
  • Server database and server code setup to parse text data and convert to MySQL. (February 29, 2016)
  • Server notifications, general logic (March 9, 2016)
  • Mobile push notifications. (March 9, 2016)

Spring Quarter Goals:

  • Finish basic notifications if necessary. (April 1, 2016)
  • Safe autonomous flying & picturing. (May 1, 2016)
  • Locating of license plate in an image using segmentation for autonomous mode. (May 1, 2016)
  • Waypointing w/ autonomous flying. (May 8, 2016)
  • Customer text notifications. (April 15, 2016)

Extra Credit:

  • Drone Feed. (May 15, 2016)
  • Query tool on app (website?) to check on current parked cars. (June 1, 2016)
  • Handling erroneous data / reading errors. (June 1, 2016)
  • Scanning to find open spots and post to website for users to check to save parking time. (June 8, 2015)

Glossary of Terms  

OCR– Optical Character Recognition, extracting text from an image.

Waypoints – DJI built in autonomous flight capability of Phantom 3 quadcopter. Drone will fly at a fixed altitude to a predetermined set of GPS coordinates.

Gimbal – Motorized camera mount used on the DJI Phantom 3 to stabilize and control pan the camera. Controllable from Android or iOS device.

HMD – (Head mounted display) device worn on the head that has a small display optic in front of one or both eyes

DJI Phantom 3 – Quadcopter Model

Requirements

User Cases/User Stories:

  • As the city’s parking enforcement, I want to be notified on my HMD when a vehicle has exceeded the allowed parking time and be able to see the licence plate/location of the offending vehicle.
    (Deadline: March 10, 2016)
  • As the city’s parking enforcement, I want to be able to see live feed on my HMD  from the drone if I wish.
    (Deadline: March 1, 2016)
  • As the city’s parking enforcement, I want to be able to query what vehicles have under a certain amount of time left to park using either my HMD or a mobile application.
    (Deadline: April 1, 2016)
  • As a user, I want to know that the OCR used on license plates is reliable and will consistently and accurately read licence plates.
    (Deadline: March 1, 2016)
  • As a user, I want to know that the parking monitoring system will only mark cars as parked past the allowable time if they truly are, so citations are not erroneously issued.
    (Deadline: March 1, 2016)
  • As a user, I want be confident that the drone will be able to safely avoid obstacles and not become an obstacle to others on roadways.
    (Deadline: June 1, 2016)
  • As a motorist, I want to be reminded of parking times, so that I do not get cited.
    (Deadline: April 1, 2016)
  • As a motorist, I want a better city parking experience, so that I don’t have to waste time trying to find a parking spot.
    (Deadline: June 1, 2016)
  • As the city’s parking enforcement, I want an easy (possibly even autonomous way) to monitor parked vehicles, so that we can enforce parking efficiently.
    (Deadline: June 1, 2016)
  • As the city’s parking enforcement, I want to know that the drone will be able to adjust for obstructions preventing it from being able to read a licence plate on a vehicle and if it cannot adjust for it, be notified.
    (Deadline: June 1, 2016)

Prototyping code (Github URL)  

https://github.com/asergian/Quadcopter

Appendices • Technologies employed

Image Recognition – OCR

DJI Phantom 3 Advanced

Google Glass