AI-Based UAVs 3-D Coverage Deployment in 6G-Enabled IoV Networks for Industry 5.0

The unmanned aerial vehicle (UAV) has been widely deployed for the emergency communication recovery in 6G-enabled Internet of Vehicles (IoV) networks for Industry 5.0 due to its advantages such as high mobility, low cost and on-demand deployment. However, in a practical three-dimensional (3D) space urban road scenario, the obstructions from the buildings or interchange bridges, as well as the limited transmit power and quantity of UAVs, cannot satisfy the vehicles’ communication quality of service (QoS) for UAV-assisted IoV networks. To address these challenges, we propose a QoS-aware UAVs 3D coverage deployment algorithm (QUCDA) to maximize the number of vehicles covered by UAVs based on the machine learning algorithms. The QUCDA takes into account the unique 3D urban environment characteristics and the QoS requirements of IoV networks, which are crucial factors that traditional optimization algorithms often overlook. More specifically, the proposed QUCDA improves on traditional genetic algorithm (GA) and consists of two key steps: first, a novel approach is developed to generate the initial population utilizing the K-means clustering algorithm to improve the quality of the solution. Subsequently, we propose the K-means initialized grey wolf optimization (KIGWO) algorithm to improve the diversity through a random replacement strategy. The experimental evaluation is conducted using a realistic 3D urban environment dataset, including the road locations, heights, and other environment details. Abundant simulation results demonstrate that, compared with the traditional GA, the particle swarm optimization (PSO) and the sine cosine algorithm (SCA), the proposed QUCDA can make the communication coverage reach 100%, effectively improve the system capacity and the coverage rate. The result is expected to be useful for achieving future smart city in 6G-enabled IoV networks for Industry 5.0.