| 动态环境感知的多目标室内路径规划方法 |
| |
| 引用本文： | 周艳,陈红,张叶廷,黄悦莹,张鹏程,杨卫军.动态环境感知的多目标室内路径规划方法[J].西南交通大学学报,2019,54(3):611-618, 632. |
| |
| 作者姓名： | 周艳 陈红 张叶廷 黄悦莹 张鹏程 杨卫军 |
| |
| 作者单位： | 电子科技大学资源与环境学院;电子科技大学大数据研究中心;武汉大学测绘遥感信息工程国家重点实验室;广州市城市规划勘测设计研究院 |
| |
| 基金项目： | 国家重点研发计划资助项目（2018YFB0505501，2016YFB0502303）；国家自然科学基金项目（41871321，41471332，41571392）；中央高校基本科研业务费专项资金资助项目（ZYGX2015J113） |
| |
| 摘 要： | 为了满足复杂室内环境中用户的多目标导航需求，提出了动态环境感知的多目标室内路径规划方法. 该方法顾及室内路径复杂度、拥挤程度与阻断事件等多维室内环境语义，扩展了节点-边表示的室内导航路网模型，通过量化表征多维室内环境语义，建立了能够综合感知室内环境语义变化的导航通行成本函数，然后，将顾及室内动态环境语义的导航通行成本函数值作为室内导航路网模型的边长，设计实现了基于Dijkstra的多目标室内路径规划算法. 通过模拟实验分析比较室内路径规划结果，实验结果表明:由于扩展后的室内导航路网模型增加了具有方向性语义的垂直组件，考虑了阻断事件因素，导航路径规划能够避开不可用连接边；在路径拥挤程度分别为轻度、缓慢和堵塞情况下，由于考虑了路径复杂度和拥挤程度，节约的通行时间平均提升了17%.
|
| 关 键 词： | 环境语义感知 路径规划 Dijkstra算法 室内导航 |
| 收稿时间： | 2018-03-06 |
Multi-objective Indoor Path Planning Method with Dynamic Environment Awareness |
| |
| ZHOU Yan,CHEN Hong,ZHANG Yeting,HUANG Yueying,ZHANG Pengcheng,YANG Weijun.Multi-objective Indoor Path Planning Method with Dynamic Environment Awareness[J].Journal of Southwest Jiaotong University,2019,54(3):611-618, 632. |
| |
| Authors: | ZHOU Yan CHEN Hong ZHANG Yeting HUANG Yueying ZHANG Pengcheng YANG Weijun |
| |
| Abstract: | A dynamic environment-aware multi-objective indoor path planning method is proposed, aimed at satisfying the multi-objective navigation requirements of users in complex indoor environments. Multi-dimensional indoor environment semantics such as indoor path complexity, the degree of congestion, and blocking events were take into account. The node-edge representation indoor navigation network model was also expanded, and a navigation traffic cost function was established by precisely quantifying the multi-dimensional indoor environment semantics. The value of the navigation traffic cost function was then taken as the side length of the model, and a multi-objective indoor path planning algorithm based on Dijkstra was designed and implemented. The results of the simulation show that navigation path planning can avoid unavailable connection edges by adding the vertical components with directional semantics and considering the blocking events factor in the extended indoor navigation network model. After the path complexity and traffic congestion were considered, the travel time is saved by an average of 17% in three traffic patterns, i.e., light, mild and heavy congestion. |
| |
| Keywords: | |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《西南交通大学学报》浏览原始摘要信息 |
|
点击此处可从《西南交通大学学报》下载全文 |
|
