空地无人集群自调节控制与动态路径规划方法  

Self⁃adaptive formation control and dynamic path planning forair⁃ground heterogeneous swarm

作　　者：赵江[1] 张璇 池沛 王英勋 ZHAO Jiang;ZHANG Xuan;CHI Pei;WANG Yingxun(School of Automation Science and Electrical Engineering,Beihang University,Beijing 100191,China;Institute of Unmanned System,Beihang University,Beijing 100191,China)

机构地区：[1]北京航空航天大学自动化科学与电气工程学院,北京100191 [2]北京航空航天大学无人系统研究院,北京100191

出　　处：《航空学报》2024年第16期207-226,共20页Acta Aeronautica et Astronautica Sinica

基　　金：中央高校基本科研业务费专项资金。

摘　　要：针对异构感知条件下空地无人集群协同避障与导航问题,提出了集群队形自调节控制与动态路径规划方法。首先,建立了无人机(UAV)与地面机器人的运动、感知与通信模型,设计了空地无人集群协调运动控制架构,满足异构集群协同避障与导航的控制目标。其次,提出了机器人集群队形边界动态生成方法,分别设计了形状控制、个体避碰、编队控制与导航控制分量,实现了基于动态边界的集群队形自调节控制。结合机器人动态队形边界特性,提出了基于动态窗口优化的无人机路径规划方法,通过设计优化目标函数在集群协同运动的基础上实现了安全导航避撞。最后,设计了队形边界缩放、变形和旋转下的队形调节场景与狭窄廊道协同避障与导航任务想定,通过仿真算例验证了空地无人集群队形自调节控制与动态路径规划方法的有效性,并给出了所述方法能力边界的分析。To deal with the collective obstacle avoidance and navigation problem of air-ground unmanned swarm with heterogeneous detection abilities,a self-adaptive formation control and dynamic path planning method is proposed in this paper.Firstly,the mathematical model of Unmanned Aerial Vehicle(UAV)and ground robots describing motion,detection and communication is established respectively.And the control architecture is designed to satisfy the target of collective obstacle avoidance and navigation of the heterogeneous swarm.Secondly,this paper proposes the dy⁃namic formation boundary generation method for ground robot swarm and designs shape control,inter-agent collision avoidance,formation and navigation control component separately to achieve self-adaptive formation control.Further⁃more,a path planning method for UAV based on dynamic window approach is proposed in view of the characteristics of dynamic formation boundary.The safety of collective navigation and obstacle avoidance is guaranteed on the basis of self-adaptive formation control by designing optimization function.Finally,the paper designs the scene of formation adaption including scaling,deformation and rotation and task situation of collective obstacle avoidance and navigation through narrow corridors.The effectiveness of the proposed self-adaptive formation control and dynamic path planning for air-ground unmanned swarm is validated through simulation,and the applicable boundary of the proposed method is further analyzed.

关 键 词：空地无人集群 无人机 地面机器人 协同避障与导航 队形自调节控制 动态路径规划 

分 类 号：V249.12[航空宇航科学与技术—飞行器设计]