Autonomous illegal parking detection system using TurtleBot4 robots, developed as part of the Doosan Robotics Bootcamp 2025 – 3rd Cohort**.
This project demonstrates a multi-robot surveillance system using two TurtleBot4 robots. The robots autonomously patrol an area, detect parked vehicles, recognize license plates, and verify them against a database. If a vehicle is departing and there is a risk of collision, or if illegal parking is detected, the robot activates a warning signal (🔊 beep-boop alert).
- A vehicle enters a blind zone and is detected using YOLO-based vision.
- The nearest TurtleBot4 robot executes Nav2 navigation to approach the vehicle.
- The robot performs local patrol and identifies any parked vehicle.
- The license plate number and current position are sent to a central system controller.
- The controller queries a local database to verify the license plate.
- If the vehicle is unauthorized → 🚨 illegal parking alert is triggered.
- The alert is visualized on a GUI dashboard and stored in a monitoring database, displaying the vehicle number and zone information.
- The system operates in two zones, each covered by a dedicated TurtleBot4 robot (multi-robot architecture).
-
robot0: Patrols Zone 0, checks vehicle presence, detects license plate, and raises alert. -
robot1: Operates in Zone 1, performs same tasks independently. -
Shared Services:
- License plate OCR module
- Vehicle number database service
- Detects vehicles using YOLOv8 (object detection) + Bytetrack (object tracking)
- Recognizes license plates using Easy OCR with depth-guided localization.
- Measures distance to license plate using a depth camera to enhance recognition accuracy.
- Implements a custom waypoint-based navigation graph.
- Employs a Breadth-First Search (BFS) algorithm for optimal path planning across patrol zones.
- Supports automatic saving and loading of waypoint configurations in YAML format, ensuring persistence and reusability.
- Monitors robot patrol status and OCR results in real time.
- Displays unauthorized vehicle detections by zone.
- Visual feedback for system alerts and mission states.
- Maintains a whitelist of authorized license plates.
- Matches OCR results against the database to determine illegal parking.
- Sends alert messages if an unauthorized vehicle is detected.
For a detailed explanation of this project, please refer to the following document:
👉 docs/F-1&2_지능1_이재호_배재성_전유진_정은영_김태영_김도엽_한준모.pdf
Thanks to these wonderful people who have contributed to this project:
 weedmo |
 yujin114 |
 ethica-alt |
jsbae-RL |
 SmiteLims |
 taeyoung730 |
 kimdoyub |