Three-Dimensional Path Planning for Unmanned Aerial Vehicles Based on a Hybrid Strategy and an Improved Dung Beetle Optimization Algorithm

In intelligent agent path planning, the dung beetle algorithm is a popular path-solving strategy that has been widely applied. This paper proposes a UAV path planning method based on an improved dung beetle algorithm to address the issues of traditional bio-inspired algorithms, such as easily falling into local optima, generating redundant waypoints, and having low convergence accuracy when solving UAV three-dimensional path planning problems. First, a mathematical model is used to establish a 3D model of mountainous terrain, taking into account the UAV's objective function and constraints to make the experimental environment more realistic. Next, Fuch and Logistic chaotic maps are introduced to initialize the dung beetle algorithm's population, enhancing its global search ability. An adaptive nonlinear decrement model is employed to dynamically adjust the number of dung beetles during the early and later stages of the algorithm, improving convergence speed. The path is then smoothed using a Bézier curve. Finally, the spiral search strategy from the sand cat optimization algorithm is incorporated into the position update of the dung beetle to improve the algorithm's local search capability and enhance convergence accuracy. The improved dung beetle optimization algorithm and other bio-inspired algorithms are applied to the UAV path planning problem. Simulation results show that the improved dung beetle optimization algorithm performs better in terms of path length and convergence speed, resulting in smoother paths, thereby validating the effectiveness of the algorithm.