面向复杂三维场景无人机自适应混沌优化航路规划算法

Adaptive Chaotic Optimization Route Planning Algorithm for UAVs in Complex Three-Dimensional Scenarios

摘要:

针对复杂三维环境下无人机航路规划对全局优化、动态避障与地形跟随的迫切需求，提出一种融合自适应机制与混沌优化的改进蚁群算法（ACACO）；通过设计随迭代过程自适应的信息素挥发因子和目标导向式启发函数，动态平衡算法的探索与开发能力；引入混沌映射对种群初始化与信息素更新扰动，有效增强算法的全局寻优能力；构建了综合航程、动态威胁与地形高程的复合代价函数；仿真结果表明，相较于传统蚁群算法（ACO）、粒子群算法（PSO）及改进雪崩算法（SAA），ACACO算法规划路径的平均长度缩短了12.3%，平滑度提升了28%，动态威胁规避成功率达到了96.7%，平均收敛迭代次数减少53.1%，在长江流域“退渔还江”司法执行案中，实现了自动化仿地勘察与动态巡查，验证其在复杂场景下的工程适用性与控制效能，为无人机在巡检、测绘等领域的智能航路规划提供了有效的算法解决方案。

Abstract:

To address the urgent demands for global optimization, dynamic obstacle avoidance and terrain following in UAV route planning within complex three-dimensional environments, an adaptive chaotic optimization ant colony algorithm (ACACO) was studied; by designing adaptive pheromone evaporation factor and goal-oriented heuristic function varying with iteration process, dynamic balance between exploration and exploitation capabilities of the algorithm was achieved; chaotic mapping was introduced to perturb population initialization and pheromone update, effectively enhancing global optimization ability; a composite cost function integrating range, dynamic threat and terrain elevation was constructed; simulation results show that compared with traditional ant colony algorithm (ACO), particle swarm optimization (PSO) and improved snow avalanche algorithm (SAA), ACACO algorithm reduces average path length by 11.1%, improves smoothness by 28%, achieves dynamic threat avoidance success rate of 96.7%, and reduces average convergence iterations by 53.1%; in the "Returning Fishing Grounds to the River" judicial enforcement case in Yangtze River basin, automatic terrain-following survey and dynamic patrol were applied, verifying engineering applicability and control effectiveness in complex scenarios; this research provides an effective algorithmic solution for intelligent route planning of UAVs in inspection, mapping and other fields.