Cooperative Sensing and Heterogeneous Information Fusion in VCPS: A Multi-agent Deep Reinforcement Learning Approach

Cooperative sensing and heterogeneous information fusion are critical to realize vehicular cyber-physical systems (VCPSs). This paper makes the first attempt to quantitatively measure the quality of VCPS by designing a new metric called Age of View (AoV). Specifically, a vehicular edge computing (VEC) architecture is presented, in which heterogeneous information can be cooperatively sensed and uploaded via Vehicle-to-Infrastructure (V2I) communications. Logical views can be constructed by fusing the heterogeneous information at edge nodes. Further, we derive a cooperative sensing model based on the multi-class M/G/1 priority queue. On this basis, we define the AoV by modeling the timeliness, completeness, and consistency of the logical views. Then, we formulate the problem, which aims at maximizing the quality of VCPS. Further, a multi-agent deep reinforcement learning solution is proposed. The system state includes vehicle sensed information, edge cached information, and view requirements. The vehicle action space consists of the sensing frequencies and uploading priorities of information, and the edge action space is the V2I bandwidth allocation. The system reward is defined as the achieved VCPS quality. In particular, a difference-reward-based credit assignment is designed to divide the system reward into the difference reward for vehicles, reflecting their individual contributions. Finally, we build the simulation model and give a comprehensive performance evaluation, which conclusively demonstrates the superiority of the proposed solution.