Digital Twin for Parking Management: A Comprehensive Perspective

Digital Twin (DT) technology has gradually emerged as a foundation in the evolution of urban mobility, particularly in the field of autonomous vehicle parking. The requirement of intelligent, responsive and scalable parking systems has grown significantly because of the increasing complexity and congestion of urban environments. This paper presents an analytical review on how 3D simulation techniques, incorporated with DT frameworks, are modelling vehicle parking processes. The core focus lies in influencing DTs for simulating real-time parking scenarios, enabling precise 3D object detection, adaptive path planning, and control strategies of self-driving vehicles. The research explores the integration of various technologies such as AI, IoT, 5G and simulation platforms like CARLA, SUMO, Unity3D and MATLAB/SIMULINK to imitate urban infrastructure in digital spaces. Advanced vision-based models such as YOLO, VGG16 and ResNet are implemented for their accuracy in object and slot detection under competitive conditions. The review also highlights hybrid ML systems that incorporate edge computing, cloud synchronization, and real-time feedback mechanisms, contributing to precise decision making and improved safety. Additionally, the paper highlights key obstacles such as dataset limitations, infrastructure rigidity, and sensor conformity, offering future directions that include mobile sensing, distributed processing, and the development of large-scale availability of 3D datasets. This comprehensive overview covers the revolutionary role of digital twins in enabling efficient parking environments, leading the way for more sustainable and technologically driven urban mobility.