基于视线的无人机鲁棒自适应编队控制设计  

Robust adaptive design for line-of-sight based on unmanned aerial vehicle formation control

作　　者：李雪松[1] 李颖晖[1] 张水清[2] 李霞[1] 徐浩军[1] 

机构地区：[1]空军工程大学工程学院,陕西西安710038 [2]北京航空工程技术研究中心,北京100076

出　　处：《飞行力学》2011年第6期31-35,共5页Flight Dynamics

基　　金：空军工程大学研究生创新基金资助

摘　　要：针对基于视线的无人机编队动力学非线性不确定的特性,设计了一种基于鲁棒自适应控制的编队控制系统。首先建立了基于视线的长机-僚机编队方式下的无人机编队动力学模型,而后将其转换成伪控制形式,将伪控制量分解成参考模型的输出、比例微分控制输出与神经网络自适应输出三项,采用误差观测器的输出训练神经网络以实现误差补偿,利用Lyapunov理论对误差的有界性进行了证明,设计了伪控制保护以防止执行器超出物理范围。最后建立了无人机六自由度非线性模型并进行仿真,结果表明设计的编队控制系统可以使僚机有效地跟踪做不确定机动的长机。Considering the nonlinearity and uncertainty of line-of-sight（LOS） based unmanned aerial vehicle（UAV） formation flight dynamics,a robust adaptive control method is proposed to design the formation flight control system.Firstly,the LOS based on formation flight control model is established.Then,the model is inverted to the ＇pseudo-controls＇ form,the pseudo-control is constructed from the outputs of reference models,the proportional-derivative controller outputs and neural network（NN） adaptive outputs,the NN which is training by the outputs of the error observer is used to compensate unmodeled dynamics,the errors are proved bounded by Lyapunov theory,the pseudo-control hedging（PCH） is proposed to prevent the actuator from rate and position limits.Finally,the simulation is accomplished based on 6-degree-of-freedom nonlinear UAV model,the simulation results show that the follower can track the leader which is flying with uncertain maneuvers well.

关 键 词：无人机编队 鲁棒自适应 误差观测器 

分 类 号：V279[航空宇航科学与技术—飞行器设计]