Platoon intelligence: edge learning in vehicle platooning networks

Abstract

Abstract With the increasing computing capacity of intelligent connected vehicles (ICVs), there will be a considerable amount of spare resources available in the vehicle cluster. It is necessary to make full use of these spare resources. In this paper, we investigate a platoon intelligence-based edge learning (PIEL) framework for the deep integration of terminal and network edge. Specifically, this paper discusses the characteristics and applications of platooning in edge learning and introduces the general architecture of PIEL. Subsequently, we conduct a discussion on the learning scheduling based on PIEL, including co-scheduling in the Intra-Platoon, Inter-Platoon, and Platoon-edge cases. An intelligent decision-making methodology framework for PIEL is also presented. Finally, we discuss future research opportunities on PIEL to achieve platoon intelligence. We believe that this investigation will elicit escalating attentions, stimulate fruitful discussions, and inspire further research ideas on edge intelligence.

Keywords

• Intelligent connected vehicle
• Intelligent transportation system
• Edge intelligence
• Multi-access edge computing