Automated Valet Parking and Charging: A Novel Collaborative AI-Empowered Architecture
Xiansheng Guo, Gordon Owusu Boateng, Haonan Si, Yu Cao, Yu Qiu, Zhexue Lai, X Li, Xinhao Liu, Cheng Chen
Abstract
The surge in global car ownership and the largescale popularization of new Electric Vehicles (EVs) have precipitated significant challenges in parking space allocation and charging infrastructure world-wide. Automated Valet Parking and Charging (AVPC) systems have emerged as the key to surmounting the aforementioned pain points. Recent advancements in Artificial Intelligence (AI) have catalyzed transformative developments in optimization strategies within AVP frameworks. This article proposes a novel layered and collaborative AI-empowered AVPC architecture for optimal and efficient automated parking and charging in parking areas. Our discourse delves into three key modules integral to the proposed architecture: High Definition (HD) map generation and updating, interactive vehicle-infrastructure collaborative sensing, and multi-vehicle global scheduling. These key modules are interwoven to facilitate seamless interaction among entities in the architecture, augmenting efficient EV parking and charging. Simulation-based evaluation analyzes the impact of the modular architecture in AVPC scenarios, demon-strating the efficiency of its key modules.