摘要
随着无人机作业空域从中高空不断向低空甚至超低空拓展,复杂的低空障碍环境对无人机造成了严重的威胁。研究无人机避障航路规划理论与方法,对于保障无人机的飞行安全和提升其任务效率具有重要作用。对无人机避障航路规划方法的研究现状进行了梳理,首先,根据航路规划问题所建立的优化模型,将规划方法划分为基于数学规划的方法、基于路标图的方法、基于空间分解的方法、基于势场的方法、基于随机规划的方法和基于机器学习的方法六个大类。然后,分别介绍了各类型方法的基本原理、代表性研究以及优缺点。最后,对避障航路规划方法未来可能的研究方向进行了展望。综述表明,复杂环境下无人机三维航路规划方法的研究仍有提升空间;未来应考虑将传统规划方法与新一代人工智能技术相结合;航路规划方法研究应充分考虑机载传感器的实际性能和工作特性;规划航路的可跟踪性问题也亟待解决。
With the extension of the unmanned aerial vehicle(UAV)operation airspace from medium or high altitude to low altitude,complex obstacle environments in low altitude seriously threaten UAVs.Investigating the theories and methodologies of UAV path planning for obstacle avoidance is essential to guarantee the flight safety and enhance the mission efficiency of UAVs.The current research status of UAV path planning methods for obstacle avoidance is combed in this review.First,according to the established optimization models in path planning problems,the planning methods are divided into mathematical programming-based approaches,roadmap-based approaches,spatial decomposition-based approaches,potential field-based approaches,stochastic programming-based approaches and machine learning-based approaches.Then,fundamental principles,representative studies,merits and demerits of each approach are introduced respectively.Finally,promising research directions of path planning for obstacle avoidance in future are prospected.The review suggests that there is still room for improvement in researches on UAV three-dimensional path planning in complex environments.In addition,we should focus more on the combination of traditional path planning methods and newgeneration artificial intelligence technologies,and take into full consideration of the actual performances and operating characteristics of onboard sensors in path planning methods.Also,the trackability of planned paths needs to be urgently solved.
作者
吴健发
王宏伦
刘一恒
姚鹏
WU Jianfa;WANG Honglun;LIU Yiheng;YAO Peng(School of Automation Science and Electrical Engineering,Beihang University,Beijing 100191,China;Science and Technology on Aircraft Control Laboratory,Beihang University,Beijing 100191,China;Shenyuan Honors College of Beihang University,Beijing 100191,China;College of Engineering,Ocean University of China,Qingdao 266100,China)
出处
《无人系统技术》
2020年第1期1-10,共10页
Unmanned Systems Technology
基金
国家自然科学基金(61175084)
山东省自然科学基金(ZR2018BF016)。
关键词
无人机
航路规划
数学规划
路标图
空间分解
人工势场
随机规划
机器学习
Unmanned Aerial Vehicles(UAVs)
Path Planning
Mathematical Programming
Roadmap
Space Decomposition
Artificial Potential Field
Stochastic Programming
Machine Learning
