动态可重构自适应无人车辆编队系统  

Dynamic Reconfigurable Adaptive UGV Formation System

作　　者：贾一飞 蒋朝阳 JIA Yifei;JIANG Chaoyang(Collective Intelligence&Collaboration Laboratory,China North Artificial Intelligence&Innovation Research Institute,Beijing 100072,China;School of Machanical Engineering,Beijing Institute of Technology,Beijing 100081,China)

机构地区：[1]中兵智能创新研究院有限公司群体协同与自主实验室,北京100072 [2]北京理工大学机械与车辆学院,北京100081

出　　处：《兵工学报》2024年第10期3654-3673,共20页Acta Armamentarii

摘　　要：为研究无人车编队系统的编队保持、队形重构及队形变换功能,提出一种混合式领航跟随策略,以降低对领航车的依赖并确保编队完整性。开发基于车间(Vehicle to Vehicle,V2V)通信的跟随车独立避障功能,并设计了实时管理编队成员属性并支持人机交互的编队节点管理系统。提出一种三维空间下的三次样条曲线动态扩展轨迹规划方法,结合V2V通信获取前车位姿信息生成跟随轨迹并实现避障。利用Frenet坐标系解耦车距保持与轨迹跟踪问题,采用PID控制器和线性二次调节(Linear Quadratic Regulator,LQR)控制器分别进行纵向控制和横向轨迹跟踪。研究结果表明:所搭建的仿真环境可快速验证方法性能,显示该方法具有良好的性能;通过实车验证了车辆编队系统的3种功能,通过车距稳定保持,证实所提方法具备良好实时性,能够实现多车编队的有效跟随,通过多种队形的变换以及成员入队离队场景,显示出高度的智能拓展性和适应性。The formation keeping,reconfiguration and transformation functions of unmanned ground vehicle(UGV)formation systems are studied A hybrid leader-follower strategy is proposed to reduce the dependence on the leading vehicle and ensure the formation integrity.An independent obstacle avoidance function based on vehicle-to-vehicle(V2V)communication for the following vehicles is developed,and a formation node management system is designed manages the attributes of formation members in real time and supports the human-computer interaction.A dynamic extended trajectory planning method with cubic spline curve in three-dimensional space is proposed to generate the following trajectory and realize the obstacle avoidance by acquiring the position information of the front vehicle through V2V communication.The Frenet coordinate system is utilized to decouple the distance keeping and trajectory tracking problems,and the proportional-integral-differential(PID)controller and linear quadratic regulator(LQR)controller are used for longitudinal control and lateral trajectory tracking,respectively.The research results show that the performance of the proposed method can be quickly verifued in the simulation environment built,showing that the method has good performance.And the three functions of the vehicle formation system are verified by real vehicles,and the proposed method is confirmed to have good real-time performance through the stable maintenance of the distance between the vehicles,which is capable of realizing the effective following of the multi-vehicle formation,and shows a high degree of intelligent expansion and adaptability through the transformation of multiple formation shapes as well as the scenarios of members joinning and departuring from the vehicle formation.

关 键 词：无人车辆 编队控制 车间通信 轨迹规划 轨迹跟踪 

分 类 号：U461[机械工程—车辆工程]