Cloud computing had emerged as a fundamental innovation in the field of Autonomous Vehicles (AV), which has been subjected to extensive research in successive years. Whereas cloud computing offers highly efficient computation, memory, and communication infrastructure, the AV still has low mobility support, poor location awareness, and significant processing latency. To solve those issues, we combine computing and Software-Defined Networking (SDN) in this study. By extending computing and storage towards the network edge, computers to enhance mobility and lower latency while also significantly reducing latency. SDN, meanwhile, gives the network adaptable centralized control and global knowledge. We develop a new SDN-based Improved Weighed Particle Swarm Optimization (SDN-IWPSO) logistic regression to effectively implement the SDN architecture in the AV. SDN-IWPSO utilizes the change of the light of particles in mutations and sequential reduce parameters to develop the efficiency of parameter Particle Swarm Optimization (PSO). According to the experimental result, the SDN-IWPSO algorithm would significantly reduce latency and raise the level of healthcare Quality of Service (QoS) inside the SDN architecture.

Autonomous Vehicle; Edge Computing; Software Defined Network; Quality of Service; Optimization