The traditional A^(*)algorithm exhibits a low efficiency in the path planning of unmanned surface vehicles(USVs).In addition,the path planned presents numerous redundant inflection waypoints,and the security is low,wh...The traditional A^(*)algorithm exhibits a low efficiency in the path planning of unmanned surface vehicles(USVs).In addition,the path planned presents numerous redundant inflection waypoints,and the security is low,which is not conducive to the control of USV and also affects navigation safety.In this paper,these problems were addressed through the following improvements.First,the path search angle and security were comprehensively considered,and a security expansion strategy of nodes based on the 5×5 neighborhood was proposed.The A^(*)algorithm search neighborhood was expanded from 3×3 to 5×5,and safe nodes were screened out for extension via the node security expansion strategy.This algorithm can also optimize path search angles while improving path security.Second,the distance from the current node to the target node was introduced into the heuristic function.The efficiency of the A^(*)algorithm was improved,and the path was smoothed using the Floyd algorithm.For the dynamic adjustment of the weight to improve the efficiency of DWA,the distance from the USV to the target point was introduced into the evaluation function of the dynamic-window approach(DWA)algorithm.Finally,combined with the local target point selection strategy,the optimized DWA algorithm was performed for local path planning.The experimental results show the smooth and safe path planned by the fusion algorithm,which can successfully avoid dynamic obstacles and is effective and feasible in path planning for USVs.展开更多
针对AGV(Automated Guided Vehicle)叉车处于环境信息未知或环境动态变化情况下的路径规划及导航问题,文中提出了一种由YOLOv5(You Only Look Once version 5)目标检测算法获取目标位置。根据目标位置规划出全局基础路径,再融合DWA(Dyna...针对AGV(Automated Guided Vehicle)叉车处于环境信息未知或环境动态变化情况下的路径规划及导航问题,文中提出了一种由YOLOv5(You Only Look Once version 5)目标检测算法获取目标位置。根据目标位置规划出全局基础路径,再融合DWA(Dynamic Window Approach)局部动态路径规划算法进行AGV路径规划与导航,使AGV叉车在未知环境或局部环境信息未知的环境中能快速识别目标位置并完成路径规划到达目标位置。实验结果表明,相较于改进前方法,文中所提方法在路径长度、耗费时间以及AGV叉车航向误差方面均有良好表现,路径长度平均减少12%,耗费时间平均减少约5%且AGV航向与目标航向的平均误差在5°以内。所提方法提高了AGV叉车在未知环境中的工作效率以及工作灵活性。展开更多
基金Supported by the EDD of China(No.80912020104)the Science and Technology Commission of Shanghai Municipality(No.22ZR1427700 and No.23692106900).
文摘The traditional A^(*)algorithm exhibits a low efficiency in the path planning of unmanned surface vehicles(USVs).In addition,the path planned presents numerous redundant inflection waypoints,and the security is low,which is not conducive to the control of USV and also affects navigation safety.In this paper,these problems were addressed through the following improvements.First,the path search angle and security were comprehensively considered,and a security expansion strategy of nodes based on the 5×5 neighborhood was proposed.The A^(*)algorithm search neighborhood was expanded from 3×3 to 5×5,and safe nodes were screened out for extension via the node security expansion strategy.This algorithm can also optimize path search angles while improving path security.Second,the distance from the current node to the target node was introduced into the heuristic function.The efficiency of the A^(*)algorithm was improved,and the path was smoothed using the Floyd algorithm.For the dynamic adjustment of the weight to improve the efficiency of DWA,the distance from the USV to the target point was introduced into the evaluation function of the dynamic-window approach(DWA)algorithm.Finally,combined with the local target point selection strategy,the optimized DWA algorithm was performed for local path planning.The experimental results show the smooth and safe path planned by the fusion algorithm,which can successfully avoid dynamic obstacles and is effective and feasible in path planning for USVs.
文摘针对AGV(Automated Guided Vehicle)叉车处于环境信息未知或环境动态变化情况下的路径规划及导航问题,文中提出了一种由YOLOv5(You Only Look Once version 5)目标检测算法获取目标位置。根据目标位置规划出全局基础路径,再融合DWA(Dynamic Window Approach)局部动态路径规划算法进行AGV路径规划与导航,使AGV叉车在未知环境或局部环境信息未知的环境中能快速识别目标位置并完成路径规划到达目标位置。实验结果表明,相较于改进前方法,文中所提方法在路径长度、耗费时间以及AGV叉车航向误差方面均有良好表现,路径长度平均减少12%,耗费时间平均减少约5%且AGV航向与目标航向的平均误差在5°以内。所提方法提高了AGV叉车在未知环境中的工作效率以及工作灵活性。
