输入受限的多旋翼无人机轨迹跟踪鲁棒正定不变集设计  

Design of robust positively invariant set for trajectory tracking of multirotor UAV with input saturation

作　　者：李志南 李卫华[1] 王鹏[1] 曹越 LI Zhi-nan;LI Wei-hua;WANG Peng;CAO Yue(Information and Navigation College,Air Force Engineering University,Xi'an 710003,China)

机构地区：[1]空军工程大学信息与导航学院,西安710003

出　　处：《控制与决策》2021年第4期857-866,共10页Control and Decision

基　　金：国家自然科学基金项目(61773396,61403414,61403297,61703423,61771484,61571458,41601436,71503260);陕西省自然科学基础研究计划项目(2016JQ6070,2017JQ6043,2018JM4029,2016JM6050);中国博士后科学基金项目(2016M603042,2017T100817);航空科学基金项目(20160196005);国防科技创新特区项目。

摘　　要：针对输入受限的多旋翼无人机轨迹跟踪问题设计了鲁棒正定不变集.考虑多旋翼无人机的非线性动力学模型、外部不确定干扰以及输入饱和约束等因素,首先设计带有干扰抵消项的非线性控制律以保证无人机轨迹跟踪的稳定性,进而基于线性矩阵不等式(LMIs)构造了鲁棒正定不变集.在所设计的控制律作用下,无人机轨迹跟踪误差一旦进入所构造的不变集则将始终处于该集合内,并最终趋于零.理论推导过程与仿真实验结果均验证了以上特性.A robust positively invariant(RPI) set is designed for the trajectory tracking problem of a multi-rotor unmanned aerial vehicles(UAVs) subject to external disturbance and input saturation. With considering the nonlinear dynamics,external disturbance and input saturation of the multi-rotor UAV, a nonlinear control law with disturbance offset is designed firstly to ensure the stability of tracking, and the RPI set is constructed based on linear matrix inequalities(LMIs). For the tracking error within the designed RPI set, the designed control law can always guarantee that the tracking error stays in the RPI set forever and eventually converges to zero. The above properties are verified by the theoretical derivation and simulation results.

关 键 词：多旋翼无人机 外部干扰 输入受限 鲁棒正定不变集 线性矩阵不等式 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]