引用本文:韩晓薇,鲜斌,杨森.无人机吊挂空运系统的自适应控制设计[J].控制理论与应用,2020,37(5):999~1006.[点击复制]
HAN Xiao-wei,XIAN Bin,YANG Sen.Adaptive controller design for an unmanned quadrotor transportation system[J].Control Theory and Technology,2020,37(5):999~1006.[点击复制]
无人机吊挂空运系统的自适应控制设计
Adaptive controller design for an unmanned quadrotor transportation system
摘要点击 1672  全文点击 878  投稿时间:2019-03-25  修订日期:2019-07-22
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2019.90181
  2020,37(5):999-1006
中文关键词  四旋翼无人机  运输系统  模型不确定性  神经网络  减摆控制
英文关键词  quadrotor unmanned aerial vehicle  transportation system  modeling uncertainties  neural network  antiswing control
基金项目  国家自然科学基金项目(91748121, 90916004, 60804004)资助
作者单位E-mail
韩晓薇 天津大学  
鲜斌* 天津大学 xbin@tju.edu.cn 
杨森 天津大学  
中文摘要
      针对四旋翼无人机吊挂空运系统存在的模型不确定性及欠驱动性问题, 本文提出了一种基于能量耦合的 自适应控制设计. 首先, 基于能量整形控制方法构造了一种新型的能量存储函数以处理状态耦合. 然后利用神经网 络对系统未建模动态特性进行在线估计, 同时设计参数自适应律在线估计模型中的未知参数, 并采用基于符号函数 的鲁棒控制算法补偿神经网络的估计误差. 本文运用李雅普诺夫方法和拉塞尔不变性原理对闭环系统的稳定性进 行了证明, 并且证明了负载摆动和无人机位置误差的渐近收敛性. 最后, 在室内实验平台上进行了飞行实验. 实验 结果表明, 本文提出的非线性控制方法能够在有效抑制吊挂负载摆动的同时, 实现无人机位置的精确控制.
英文摘要
      This paper presents an energy coupling-based adaptive control scheme for an unmanned quadrotor transportation system which is subjected to modeling uncertainties and underactuated properties. A new storage function is constructed based on energy shaping methodology to deal with dynamic states coupling of the unmanned quadrotor transportation system. A neural network (NN) is used to estimate the time-varying modeling uncertainties with on-line weight tunning. And an adaptive nonlinear control law is developed to compensate for unknown parameters in the dynamics model, while the NN approximation errors are compensated by using the signum function. Lyapunov-based stability analysis and the LaSalle invariance theorem are employed to the stability of the closed-loop system, and the asymptotic convergence of the payload’s swing motion and the quadrotor’s position errors. Finally, real-time flight experiments are performed on a self-build indoor unmanned aerial vehicle (UAV) testbed. The experimental results are included to demonstrate the effectiveness of the proposed control law.