Design of a Demonstration System for Vehicular Edge Computing Scheduling

With the widespread adoption of intelligent in-vehicle devices, demands for server computing performance and network transmission speed have significantly increased, while resources in vehicle-edge computing remain limited. This paper presents a demonstration and learning platform for vehicular edge computing scheduling, employing Web3D technology to create a three-tier model of the architecture, vividly illustrating the resource allocation process. Addressing the dynamic complexity of in-vehicle network environments, this study abandons conventional scheduling methods, opting instead for deep learning algorithms to determine data offloading strategies that can adapt to changing conditions. Furthermore, the design incorporates a virtual academy where users can learn about edge computing, deep learning, and related subjects to gain a comprehensive understanding of the demo system.