基于权重控制剪枝的无人机服务替换方法

UAV Service Replacement Method Based on Weight Control Pruning

无人机执行飞行任务期间具有不稳定性,针对无人机节点崩溃后确保任务完成率的问题,依据崩溃任务的组合服务替换流程,生成替补无人机组合服务节点图,采用权重控制剪枝算法,以两步剪枝的方式对具有多维度资源的无人机节点进行处理以减少最优资源映射路径搜索复杂度,完成组合服务图中最佳无人机服务路径的快速查找,实现节点的最佳替换以及任务的自适应动态调整,以提高无人机协作效率和任务完成率。在不同路径数量的无人机替补服务网以及不同连接情况下验证了权重控制剪枝算法的有效性,可以有效地提高无人机备选方案的查找效率,完成节点的快速选择替换。

UAVs have unstable characteristics during flight task.To ensure the task completion rate after UAV node crash,the composite service node graph of replacement UAVs is generated according to the composite service replacement process of the crash task.Then weight control pruning algorithm is used.The UAV nodes with multi-dimensional resources are processed by two-step pruning to reduce the complexity of searching the optimal resource mapping path,accomplish the fast search of the best UAV service path in the composite service graph,and realize the optimal replacement of nodes and adaptive dynamic adjustment of tasks,so as to improve the UAVs cooperation efficiency and task completion rate.The effectiveness of the weight control pruning algorithm is verified in UAV service replacement network with different number of paths and different connections.It can effectively improve the search efficiency of UAV alternatives and complete the rapid selection and replacement of UAV nodes.