Multi-UAVs assisted path planning method for terrain-oriented air-ground collaborative vehicular network architecture

The application of air-ground collaborative network has become increasingly widespread in intelligent vehicular systems. In order to effectively utilize multiple unmanned aerial vehicles (UAVs) to provide fast services and improve resource allocation for air-ground vehicular network, this paper proposes a 3D terrain-oriented path planning algorithm for multi-UAVs assisted intelligent vehicular network based on swarm intelligence optimization. It is aimed to address UAVs' air-ground path planning in complex 3D terrains, requiring substantial computation. However, the current multi-objective bald eagle search algorithm tends to approach the center point, resulting in low accuracy when solving such problems. Firstly, the 3D terrain environment model, threat source model, and other models were constructed, and the multi-objective cost function was determined. Secondly, a coupled chaotic map initialization was designed to effectively improve the quality of the initialized population. In addition, an adaptive Gaussian walk strategy based on the "reconnaissance eagle" was designed to balance development and search capabilities. The fast non-dominated sorting was introduced to further improve the algorithmic efficiency. Finally, we utilized the correlation between the position of the bald eagle and UAV flight parameters of speed, turning angle, and climbing angle. An improved multi-objective bald eagle search (IMBES) was designed to efficiently search for UAV configuration space and find the optimal Pareto front. The experimental results show that the designed IMBES algorithm has a success rate of 70.50%. This proposed method offers improved optimization ability, smoother paths, and reduced energy consumption for optimizing collaborative terrain-oriented air-ground path planning, compared with existing path planning methods.