针对动态不确定战场环境下多无人机对多区域、多目标的协同察打任务规划过程中存在的信息不确定、任务多约束及航迹强耦合的多目标优化与决策问题,结合Dubins航迹规划算法,提出了一种融合多种改进策略的灰狼优化算法(grey wolf optimiza...针对动态不确定战场环境下多无人机对多区域、多目标的协同察打任务规划过程中存在的信息不确定、任务多约束及航迹强耦合的多目标优化与决策问题,结合Dubins航迹规划算法,提出了一种融合多种改进策略的灰狼优化算法(grey wolf optimization algorithm incorporating multiple improvement strategies,IMISGWO).首先,针对动态环境带来的无人机巡航速度及察打任务消失时间的不确定性,基于可信性理论建立了以最大化任务收益为指标的任务规划数学模型;其次,为实现该问题的快速求解,设计了初始解均匀分布、个体通信机制调整、动态权重更新和跳出局部最优等策略,提升算法解搜索能力;最后,构建了多无人机察打一体典型任务仿真场景,通过数字仿真以及虚实结合半实物仿真试验验证了算法的可行性和有效性.仿真结果表明:算法在求解不确定环境下耦合航迹的多无人机察打一体任务规划问题时,能够生成多机高效的任务执行序列和满足无人机飞行性能约束的飞行轨迹,且能够适用于无人机数量增加导致问题复杂度增加情形下此类问题的求解.展开更多
针对雨雾等复杂天气下无人机图像质量下降导致目标检测效果不佳的问题,提出基于上下文引导和提示学习的目标检测算法CGP-YOLO(context-guided and prompt-based YOLOv8)。构建一个多任务联合学习的检测网络,通过双分支结构达到平衡图像...针对雨雾等复杂天气下无人机图像质量下降导致目标检测效果不佳的问题,提出基于上下文引导和提示学习的目标检测算法CGP-YOLO(context-guided and prompt-based YOLOv8)。构建一个多任务联合学习的检测网络,通过双分支结构达到平衡图像检测和恢复的任务。提出基于提示学习的跨层注意力加权图像去噪分支,指导网络利用退化提示重构清晰的图像;模型主干设计基于上下文的残差采样模块,集成卷积注意力机制,综合目标的局部和全局信息;采用可分离大核多尺度特征提取模块,处理网络多尺度特征;引入小目标的专用检测头,增强小目标的检测精度。实验结果表明,在参数量仅为基线模型60%的情况下,该模型的检测精度提高了2.4个百分点,平均精度(mAP)提高了2.04个百分点,模型检测效果优于其他经典模型,具备卓越的性能。展开更多
Complex multi-area collaborative coverage path planning in dynamic environments poses a significant challenge for multi-fixed-wing UAVs(multi-UAV).This study establishes a comprehensive framework that incorporates UAV...Complex multi-area collaborative coverage path planning in dynamic environments poses a significant challenge for multi-fixed-wing UAVs(multi-UAV).This study establishes a comprehensive framework that incorporates UAV capabilities,terrain,complex areas,and mission dynamics.A novel dynamic collaborative path planning algorithm is introduced,designed to ensure complete coverage of designated areas.This algorithm meticulously optimizes the operation,entry,and transition paths for each UAV,while also establishing evaluation metrics to refine coverage sequences for each area.Additionally,a three-dimensional path is computed utilizing an altitude descent method,effectively integrating twodimensional coverage paths with altitude constraints.The efficacy of the proposed approach is validated through digital simulations and mixed-reality semi-physical experiments across a variety of dynamic scenarios,including both single-area and multi-area coverage by multi-UAV.Results show that the coverage paths generated by this method significantly reduce both computation time and path length,providing a reliable solution for dynamic multi-UAV mission planning in semi-physical environments.展开更多
文摘针对雨雾等复杂天气下无人机图像质量下降导致目标检测效果不佳的问题,提出基于上下文引导和提示学习的目标检测算法CGP-YOLO(context-guided and prompt-based YOLOv8)。构建一个多任务联合学习的检测网络,通过双分支结构达到平衡图像检测和恢复的任务。提出基于提示学习的跨层注意力加权图像去噪分支,指导网络利用退化提示重构清晰的图像;模型主干设计基于上下文的残差采样模块,集成卷积注意力机制,综合目标的局部和全局信息;采用可分离大核多尺度特征提取模块,处理网络多尺度特征;引入小目标的专用检测头,增强小目标的检测精度。实验结果表明,在参数量仅为基线模型60%的情况下,该模型的检测精度提高了2.4个百分点,平均精度(mAP)提高了2.04个百分点,模型检测效果优于其他经典模型,具备卓越的性能。
基金National Natural Science Foundation of China(Grant No.52472417)to provide fund for conducting experiments.
文摘Complex multi-area collaborative coverage path planning in dynamic environments poses a significant challenge for multi-fixed-wing UAVs(multi-UAV).This study establishes a comprehensive framework that incorporates UAV capabilities,terrain,complex areas,and mission dynamics.A novel dynamic collaborative path planning algorithm is introduced,designed to ensure complete coverage of designated areas.This algorithm meticulously optimizes the operation,entry,and transition paths for each UAV,while also establishing evaluation metrics to refine coverage sequences for each area.Additionally,a three-dimensional path is computed utilizing an altitude descent method,effectively integrating twodimensional coverage paths with altitude constraints.The efficacy of the proposed approach is validated through digital simulations and mixed-reality semi-physical experiments across a variety of dynamic scenarios,including both single-area and multi-area coverage by multi-UAV.Results show that the coverage paths generated by this method significantly reduce both computation time and path length,providing a reliable solution for dynamic multi-UAV mission planning in semi-physical environments.
