融合改进A*与增强型DWA算法下的无人路面洒水车分层路径优化方法

A Hierarchical Path Optimization Method for Unmanned Road Sprinkler Vehicles Integrating Enhanced A* and Advanced DWA Algorithms

摘要:

为了解决无人路面洒水车应用传统A*算法在路径规划中存在节点冗余、转角过大、搜素陷入局部最优等问题，提出一种融合改进A*与增强型DWA算法下的分层路径优化方法；利用基于夹角约束的扩展邻域搜索策略，减少节点访问量和转角度数；通过稠密分层对启发函数进行优化，增强路径搜索的目的性；采用基于安全关键点的路径平滑策略，减少转折点数目；并将改进的A*算法的路径关键点作为增强型DWA的中间目标点，做全局引导，以自适应环境变化，从而提高无人车动态避障灵敏度；经MATLAB仿真与实车测试实验，验证了同等条件下，所提改进融合导航算法，相较于传统导航算法以及文献[20]融合导航算法，在能够有效避开未知动静态障碍物，工作效率与安全性显著提升，从而为无人车应用领域提供有价值的解决方案。

Abstract:

To address the problems of node redundancy, excessive turning angles, and getting trapped in local optima in the path planning of unmanned road surface water sprinklers using the traditional A* algorithm, a hierarchical path optimization method integrating an improved A* algorithm and an enhanced DWA algorithm is proposed. By em-ploying an extended neighborhood search strategy based on angle constraints, the number of node visits and turning angles are reduced. The heuristic function is optimized through dense layering to enhance the purposefulness of path search. A path smoothing strategy based on safety key points is adopted to reduce the number of turning points. The key points of the improved A* algorithm's path are used as intermediate target points for the enhanced DWA algorithm to provide global guidance and adapt to environmental changes, thereby improving the dynamic obstacle avoidance sensitivity of unmanned vehicles. Through MATLAB simulation and real vehicle testing experiments, it is verified that under the same conditions, the proposed improved fusion navigation algorithm, compared with the traditional navigation algorithm and the fusion navigation algorithm in reference [20], can effectively avoid un-known dynamic and static obstacles, significantly improving work efficiency and safety, thus providing a valuable solution for the application field of unmanned vehicles.