航空学报 > 2024, Vol. 45 Issue (1): 629233-629233   doi: 10.7527/S1000-6893.2023.29233

全驱系统理论及其在航空航天领域的应用专栏

固定翼无人机紧密编队的鲁棒协同跟踪控制

张清瑞, 刘赟韵, 孙慧杰, 朱波()   

  1. 中山大学 航空航天学院,深圳 518107
  • 收稿日期:2023-06-28 修回日期:2023-09-04 接受日期:2023-10-07 出版日期:2024-01-15 发布日期:2023-10-24
  • 通讯作者: 朱波 E-mail:zhubo5@mail.sysu.edu.cn
  • 基金资助:
    国家自然科学基金(62103451)

Robust cooperative tracking control for close formation of fixed⁃wing unmanned aerial vehicles

Qingrui ZHANG, Yunyun LIU, Huijie SUN, Bo ZHU()   

  1. College of Aeronautics and Astronautics,Sun Yat?sen University,Shenzhen 518107,China
  • Received:2023-06-28 Revised:2023-09-04 Accepted:2023-10-07 Online:2024-01-15 Published:2023-10-24
  • Contact: Bo ZHU E-mail:zhubo5@mail.sysu.edu.cn
  • Supported by:
    National Natural Science Foundation of China(62103451)

摘要:

固定翼无人机(UAV)具有典型的欠驱动、非线性等特点,导致航点自主跟踪设计难度大,同时紧密编队飞行过程中无人机之间运动耦合与气动干扰明显,进一步增加了高性能控制设计难度。为此,本文针对固定翼无人机紧密编队飞行的航点自主跟踪问题,综合考虑轨迹平滑性、运动协同性和跟踪鲁棒性等需求,依托全驱系统建模方法提出了一种多层次鲁棒协同跟踪控制架构。该架构包括上层运动规划、中层协同滤波及底层鲁棒协同跟踪控制3个核心部分。上层运动规划根据离散航点指令,融合虚拟结构法和迭代线性二次型优化,实时生成可行、平滑的运动轨迹。在此基础上,为改善跟踪控制的瞬态性能,引入分布式协同滤波思想,对规划轨迹进行滤波处理,生成每架无人机个体的参考信号。最后,考虑到紧密编队中气动耦合与系统不确定性等挑战,设计了一种基于不确定性及干扰观测器的鲁棒协同跟踪控制方法,实现了编队轨迹的精确跟踪以及队形的可靠保持。所提出的鲁棒协同跟踪控制架构考虑了紧密编队飞行中的多种约束与挑战,实现了编队运动规划与跟踪控制的综合设计,可以有效提升系统自主性、协同性和鲁棒性。最后,通过5架固定翼无人机的紧密编队飞行仿真对所提方法进行综合测试,验证了方法的有效性。

关键词: 协同控制, 鲁棒控制, 紧密编队控制, 固定翼无人机, 多无人机系统, 运动规划

Abstract:

Fixed-wing Unmanned Aerial Vehicle (UAV) has typical characteristics such as underactuation and nonlinearity, which poses challenges in autonomous waypoint tracking design. Additionally, during close formation flight, there is significant motion coupling and aerodynamic interference among unmanned aerial vehicles, further complicating high-performance control design. To address the autonomous waypoint tracking problem in close formation flight of fixed-wing unmanned aerial vehicles, considering requirements for trajectory smoothness, motion coordination, and tracking robustness, a multi-level robust cooperative tracking control architecture is proposed based on the fully-actuated system modeling method. This architecture consists of three core components: motion planning at upper-level, cooperative filtering at middle-level, and robust cooperative tracking control at lower-level. The upper-level generates feasible and smooth motion trajectories in real-time by integrating the virtual structure method with iterative linear quadratic regulator based on discrete waypoint. To improve the transient performance of tracking control, distributed cooperative filtering is introduced to filter the planned trajectories, generating reference signals for each individual unmanned aerial vehicle. Finally, considering challenges such as aerodynamic coupling and system uncertainties in close formation, a robust cooperative tracking control method based on uncertainty and disturbance observers is designed, achieving precise trajectory tracking and reliable formation maintenance. The proposed robust cooperative tracking control architecture addresses various constraints and challenges in close formation flight, realizing the comprehensive design of formation motion planning and tracking control, which effectively enhances system autonomy, coordination, and robustness. Finally, test of the proposed method is conducted through simulation of close formation flight with five fixed-wing unmanned aerial vehicles, validating the effectiveness of the approach.

Key words: cooperative control, robust control, close formation control, fixed-wing unmanned aerial vehicle, multiple unmanned aerial vehicles system, motion planning

中图分类号: