摘要
储罐探伤爬壁机器人全遍历路径规划要求机器人高效遍历储罐外壁全部无障碍物区域。结合邻接矩阵、路径选择函数,提出一种单元分解算法并应用于爬壁机器人路径规划中。首先,将爬壁机器人的工作环境简化为二维平面,并通过分析漏检面积确定机器人遍历基本路径;接着,建立栅格环境,对每个栅格赋予xi值以表示其栅格状态;其次,采用矩形分解法将工作环境划分为若干子区域,通过图的深度优先搜索算法和邻接矩阵确定各子区域的衔接顺序;最后,在子区域的遍历和切换过程中,引入方向函数yi来判断爬壁机器人是否陷入死区,结合xi值提出路径选择函数fi以引导爬壁机器人快速逃离死区。在虚拟环境中进行了仿真实验,仿真结果表明,该方法不仅能引导爬壁机器人以高覆盖率和低重复率遍历工作区域,而且能快速地逃离死区。全遍历路径规划的实现拓展了爬壁机器人在检测储罐罐壁中的应用。
The oil tank inspection wall climbing robot is required to traverse all the barrier free ar-eas on the outer wall of the tank efficiently in complete coverage path planning .Combining with adjacency matrix and path selection function a grid decomposition algorithm was proposed for path planning of wall climbing robot .Firstly,the working environment of the wall climbing robot was simplified into two-dimensional planes and the traversed basic path of robot was determined by analyzing the missed area .Secondly , the grid environment was established and each grid was given a x value to represent its state . Thirdly,the work environment was divided into several sub-regions by rectangle decomposition method . The connection order of each sub-region was de-termined by the depth-first search algorithm and the adjacency matrix . Finally , in order to deter-mine whether or not the robot was stuck into the dead zone during the traversal and handover of the sub-region a path selection function fi was proposed by combining with the x i value which could guide the robot to escape from the dead zone quickly .The proposed method was simulated in the virtual environment .The simulation results showed that this method could not only guide the wall climbing robot to achieve complete coverage of the working area with a high coverage rate and a low repetition rate,but also quickly escape from the dead zone .The realization of com-plete coverage path planning can extend the application of wall climbing robot in detecting tank walls .
作者
唐东林
袁波
胡琳
李茂扬
魏子兵
TANG Dong-lin;YUAN Bo;HU Lin;LI Mao-yang;WEI Zi-bing(School of Mechatronic Engineering,Southwest Petroleum University,Chengdu 610500,China;Key Laboratory of Oil ＆ Gas Equipment,Ministry of Education,Southwest Petroleum University,Chengdu 610500,China)
出处
《工程设计学报》
CSCD
北大核心
2018年第3期253-261,共9页
Chinese Journal of Engineering Design
基金
四川省科技支撑计划资助项目(2017FZ0033)
关键词
爬壁机器人
全遍历路径规划
深度优先搜索算法
邻接矩阵
方向函数
路径选择函数
wall climbing robot
complete coverage path planning
depth first search algorithm
adjacency matrix
direction function
path selection function
