Collaborative Perception and Collaborative Computing for Value Maximization in Smart Grid

With the integration and development of smart grid and vehicle networking, autonomous vehicles make the power grid system and vehicles fully perceive the urban environment and load changes through collaborative perception and computing technology, and carry out power grid management tasks and driving tasks without manual intervention. However, collaborative perception necessitates the sharing of a substantial amount of sensing data, and collaborative computing makes it difficult to consider the interests of each participant simultaneously. Therefore, in this paper, we establish a value evaluation system for sensing block messages to perform collaborative perception and computing among vehicles in the smart grid, which can achieve mutual benefit and win-win situations between vehicles and comprehensive perception of the grid environment. Next, we jointly optimize the allocation of sensing message blocks, communication and computing resources, and formulate a sum value maximization problem for the target vehicle with time delay constraints. Given that the above problem is a mixed-integer nonlinear programming (MINLP) problem, we adopt linear relaxation and a successive convex approximation (SCA) algorithm to solve it. Finally, the simulation results demonstrate the superiority of the proposed scheme compared with existing baseline schemes.