无人机碰撞规避路径规划算法研究  

A Study on Path Planning Algorithms of UAV Collision Avoidance

作　　者：徐钊 胡劲文[2] 马云红[1] 王曼 赵春晖[2] XU Zhao;HU Jinwen;MA Yunhong;WANG Man;ZHAO Chunhui(School of Electronics and Information, Northwestern Polytechnical University, Xi′an 710072, China;School of Automation, Northwestern Polytechnical University, Xi′an 710072, China)

机构地区：[1]西北工业大学电子信息学院,陕西西安710072 [2]西北工业大学自动化学院,陕西西安710072

出　　处：《西北工业大学学报》2019年第1期100-106,共7页Journal of Northwestern Polytechnical University

基　　金：国家自然科学基金(61803309;61603303);中国博士后基金(2017M610650;2018M633574;2018T111096);爱生创新发展基金(ASN-IF2015-1502);中央高校科研业务基本费(3102017JG02011)资助

摘　　要：无人机(unmanned aerial vehicle,UAV)技术是目前国内外的研究热点。无人机系统正向着智能化、自主化的方向发展,其中路径规划是无人机自主控制的重要组成部分及无人机飞行安全的重要保障。为优化无人机障碍规避路径规划算法,分别设立静态障碍物和动态障碍物环境,基于最小规避距离和航程比这2个指标,比较分析了人工势场法、模糊逻辑算法和蚁群算法对无人机碰撞规避路径规划的性能,并针对人工势场法易陷入局部极小值的缺陷提出了通过增加垂直引导斥力来使无人机逃离局部极小值的改进措施,实验仿真严谨可靠,为进一步融合多种算法、优化现有路径规划算法奠定了基础。The unmanned aerial vehicle (UAV) has been a research hotspot worldwide. The UAV system is developing to be more and more intelligent and autonomous. UAV path planning is an important part of UAV autonomous control and the important guarantee of UAV′s safety. For the purpose of improving the collision avoidance and path planning algorithms, the artificial potential field, fuzzy logic algorithm and ant colony algorithm are simulated respectively in the static obstacle and dynamic obstacle environments, and compared based on the minimum avoidance distance and range ratio. Meanwhile, an improved algorithm of artificial potential field is proposed, and the improvement helps the UAV escape the local minimum by introducing the vertical guidance repulsion. The simulation results are rigorous and reliable, which lay a foundation for the further fusion of multiple algorithms and improving the path planning algorithms.

关 键 词：无人机 蚁群算法 模糊逻辑 人工势场法 路径规划 碰撞规避 

分 类 号：TG156[金属学及工艺—热处理]