Abstract

This work develops a mobility- and energy-efficient data routing protocol for UAV-assisted vehicular ad hoc networks (VANETs). In the proposed system, one UAV acts as a flying RSU (or roadside unit) that collects data from vehicles. At the same time, the remaining UAVs are treated as relay nodes, relaying information to the mobility service center (MSC). The UAVs can change their 3D positions within a designated area (box) at any time to ensure stable communication. The main goal of the protocol is to be conservatively active with UAV movements and data transfer, thereby balancing the routing task across the available system resources and the remaining energy levels of the UAVs. The problem is formulated as a mixed-integer nonlinear programming (MINLP) model that optimizes routing paths for UAVs and their 3D positions during data transfers. Recognizing that the problem is non-convex, a hybrid methodology is used to address it, particularly for the routing task using Integer Linear Programming (ILP) and for optimizing UAV positions using Particle Swarm Optimization (PSO) metaheuristics. The method was compared with other swarm-intelligence meta-heuristics and demonstrated good performance across all scenarios in the numerical simulations.