林区步行最优路径分析方法  

Optimal Walking Path Analysis Method for Forest Region

作　　者：陈玥璐 赵天忠[1] 武刚[1] 陈飞翔[1] CHEN Yuelu;ZHAO Tianzhong;WU Gang;CHEN Feixiang(School of Information Science and Technology, Beijing Forestry University, Beijing 100083, China)

机构地区：[1]北京林业大学信息学院,北京100083

出　　处：《农业机械学报》2018年第6期198-206,共9页Transactions of the Chinese Society for Agricultural Machinery

基　　金：国家重点研发计划项目(2017YFD0600906)

摘　　要：针对林区步行环境的复杂性与寻路目标的多样性,提出了林区步行最优路径分析方法。首先,引入了可步行性的概念,建立综合可步行性栅格,将特定的林区步行最优路径分析问题转化为避让不可行域、在可行域与易行域中计算多目标最小成本路径的问题;随后,将步行成本分为地形成本与地物成本,分别计算单因素成本并加权组合为综合步行成本,将多目标优化问题简化为单目标问题进行求解;最后,基于综合可步行性栅格,扩展了邻接表的数据结构,利用8方向邻域模式建立栅格单元间的邻接关系并计算步行成本,构建栅格网络模型。采用Dijkstra算法实现林区步行最优路径的求解。试验结果表明:本方法能够融合开放空间和离散路网,对林区步行环境进行建模,同时降低多个寻路目标的路径成本,使结果偏重于主要目标,具有一定的适用性。In forest region, walking environment is hybrid space mixed road networks and open space. People usually have diverse wayfinding objectives. In order to compute optimal walking path and provide wayfinding decision support, an optimal walking path analysis method for forest region （ OWPAM - FR） was proposed. Firstly, the concept of walkability was introduced to build comprehensive walkability raster （CWR）. Walking experiment in forest region was divided into unwalkable area, walkable area and easyto-walk area. The special optimal walking path analysis problem can be transformed into a general multiobjectives least-cost path analysis problem in walkable area and easy-to-walk area. Secondly, walking costs were divided into terrain costs, including horizontal distance, spatial distance, slope, time and energy, and feature costs, including land cover and cognitive load. Single walking costs were computed and weighted, which were combined with comprehensive walking costs. The multi-objectives problem can be simplified as the single objective problem. Thirdly, adjacency list was extended and queen＇ s pattern was used to establish adjacency relationships of raster cells based on CWR. Then the comprehensive walking costs were computed and the raster network model was successfully established. Dijkstra algorithm was used to compute the optimal walking path. The results showed that OWPAM - FR can effectively model walking environment of forest region mixed open space and road networks. At the same time, the proposed method was able to reduce path cost of muhiple wayfinding objectives and emphasis on different main objectives. In addition, OWPAM - FR composed of modeled steps had a certain applicability, which could be extended to several types of optimal walking path analysis applications, such as forest tourism, disaster relief and field investigation.

关 键 词：林区 可步行性 步行最优路径 多目标 步行成本 栅格网络模型 

分 类 号：S77[农业科学—森林工程] P208[农业科学—林学]