Fuzzy Logic based Autonomous Parking Systems - Part I: An Integrated Multi-functional System

Yu Wang, Xiaoxi ZhuarXiv 2019 · OpenReview

TL;DR

This paper introduces an intelligent autonomous parking system that integrates onboard fuzzy logic control for parking maneuvers with a remote control center for monitoring and management. The system goes beyond simple parking by incorporating automatic slot detection and supporting both parallel and vertical parking modes, aiming to enhance convenience and safety in parking scenarios.

What Problem Does This Paper Address?

The increasing number of vehicles has made finding parking spaces a time-consuming and often frustrating experience, leading to minor collisions and scratches. While autonomous parking systems exist, they often have a narrow focus, typically performing only the parking maneuver itself and relying on human observation to find a suitable spot. This paper addresses the need for a more comprehensive autonomous parking solution that not only handles the parking process but also includes automatic parking slot detection and a centralized management system for enhanced safety and efficiency.

Methods

The proposed system comprises two main sub-systems: the Fuzzy-Based Onboard System (FBOS) and a control center.

  • Fuzzy-Based Onboard System (FBOS): This system is the "brain" of the autonomous car. It utilizes a microprocessor, fuzzy logic controllers, sensors (infra-red proximity sensors, ArduIMU, rotary encoder, motion sensor), and an RF communication module. The FBOS is responsible for:
  • Slot Detection: Using infra-red sensors (FL, BL, FR, BR) to measure distances and a rotary encoder to estimate slot dimensions (width and length).
  • Parking Control: Employing fuzzy logic controllers for:
  • Posture Stabilization: Maintaining a safe distance from walls during maneuvers.
  • Steering Control: Achieving precise turning angles, especially for the 90-degree turns required in vertical parking, using a fuzzy-based PID controller.
  • Parking Decision-Making: Determining whether to perform parallel or vertical parking based on detected slot dimensions.
  • Autonomous Driving Modes: Including searching for a slot, parking, sleeping (idle), and leaving the parking spot.
  • Emergency Handling: Utilizing motion sensors for crash detection and obstacle avoidance.
  • Control Center: This system provides remote oversight and management. It consists of:
  • Ground Station: Located in the car park, it communicates with the car via real-time RF.
  • Graphic User Interface (GUI): In the control room, it displays car park status, parking progress, and allows communication with car owners via SMS (using Bluetooth and a GSM modem).
  • Communication: Real-time duplex RF communication between the car and ground station, and RS232 serial port communication between the ground station and the GUI.

Datasets

Not discussed in the paper. The authors mention constructing a model car park and a car prototype for experimental testing.

Key Insights and Ideas

  • Integrated Approach: The paper highlights the benefit of integrating slot detection with the parking maneuver, moving beyond systems that only perform the latter.
  • Fuzzy Logic for Uncertainty: The use of fuzzy logic controllers is justified by their inherent superiority in handling the uncertainties and imprecision common in real-world parking scenarios, mimicking human reasoning.
  • Dual Parking Modes: The system's ability to handle both parallel and vertical parking based on slot geometry is a significant practical advantage.
  • Remote Monitoring and Safety: The control center adds a layer of safety and convenience by allowing real-time monitoring of the car park and enabling communication with car owners.
  • Modular Design: The system's division into FBOS and a control center allows for scalability and potential commercialization.

Key Results

The paper demonstrates the feasibility of the designed intelligent autonomous parking system through experiments on a 1:14 scale car prototype and a model car park. The experiments successfully tested:

  • Auto-driving capabilities.
  • Slot detection for both parallel and vertical parking.
  • The execution of parallel and vertical parking maneuvers.
  • Emergency handling, including obstacle avoidance and congestion reporting.
  • Communication between sub-systems (RF, RS232, Bluetooth, SMS).

The experiments showed that the FBOS was robust in selecting appropriate parking modes and executing maneuvers. The GUI successfully displayed car status and location, and the communication channels were functional.

Limitations and Future Work

The authors acknowledge several limitations and suggest future improvements:

  • Steering Angle Control: The current implementation has limited steering angle control (two values: small and large) due to frictional forces and other noises. Future work should explore more advanced methods and components for finer steering angle control to improve performance.
  • Fine-tuning: The ideal steering angle can be set to any value, but fine-tuning is currently prohibited.
  • Future Improvements: Focus on improving steering angle control is explicitly mentioned as a key area for future development.

Summary updated

Abstract

This paper presents an intelligent autonomous parking system with multi-functions. The integrated system consists of two sub-systems, namely the Fuzzy-Based Onboard System and control center. Unlike most current auto-parking experiments, this FBOS enables the car to perform both slot detection and parking under two different parking modes. Facilitated by the control center, real-time monitoring is also achieved to reduce chances of error.
BibTeX
@misc{wang2019fuzzy,
  author = {Yu Wang and Xiaoxi Zhu},
  title = {{Fuzzy Logic based Autonomous Parking Systems - Part I: An Integrated Multi-functional System}},
  year = {2019},
  eprint = {1912.06764},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1912.06764},
}
Figure