Ubiquitous Control Over Heterogeneous Vehicles: A Digital Twin Empowered Edge AI Approach

Article Computer Science, Hardware & Architecture

Digital Twin Empowered Mobile Edge Computing for Intelligent Vehicular Lane-Changing

Bo Fan et al.

Summary: This article proposes a Digital Twin empowered mobile edge computing architecture to address the challenges faced by Connected and Automated Vehicles (CAVs) during lane-changing. Through this architecture, CAVs can learn in real-time and make 'foresight intelligent' lane-changing decisions to improve safety and traffic flow efficiency. By training lane-changing strategies using deep reinforcement learning, the effectiveness of this architecture is validated through results.

IEEE NETWORK (2021)

添加到收藏夹

Article Engineering, Civil

Intelligent Edge Computing in Internet of Vehicles: A Joint Computation Offloading and Caching Solution

Zhaolong Ning et al.

Summary: The Internet of Vehicles (IoV) field requires intelligent offloading strategies and efficient decision-making solutions, and artificial intelligence (AI) and machine learning technologies can enhance the intelligence and performance of IoVs. By utilizing Mixed Integer Non-Linear Programming and an online multi-decision making scheme, it is possible to effectively reduce network delays and achieve near-optimal performance.

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2021)

添加到收藏夹

Article Computer Science, Hardware & Architecture

Extensive Edge Intelligence for Future Vehicular Networks in 6G

Weijing Qi et al.

Summary: The 6th generation mobile network (6G) is expected to achieve a fully connected world with edge intelligence (EI) as a key enabling technology, combining artificial intelligence (AI) with mobile edge computing (MEC) to release the potential of edge-side data. However, additional edge resources from parked vehicles are introduced to address the insufficiency of edge resources from roadside units (RSUs) and base stations for future vehicular networks. An intelligent service-oriented edge resource management architecture is proposed to coordinate extensive edge resources and EI-assisted services, utilizing knowledge extracted from sensory data in vehicular networks.

IEEE WIRELESS COMMUNICATIONS (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic