小型无人机水平导航控制研究

在考虑小型无人机特点和多种导航方法的基础上,研究了利用ＧＰＳ／航路推算组合导航方法针对无人机进行水平导航控制的方案,并针对一种小型无尾无人机进行了导航控制规律设计,仿真结果表明,这种组合导航方法即简单又经济,精度高、实时性强,可靠性好,可以构成具有较强实用性的自主式导航系统。     

A Hazard Analysis-based Approach to Improve the Landing Safety of a BWB Remotely Piloted Vehicle

The BUAA-BWB remotely piloted vehicle (RPV) designed by our research team encountered an unexpected landing safety problem in flight tests. It has obviously affected further research project for blended-wing-body (BWB) aircraft configuration characteristics. Searching for a safety improvement is an urgent requirement in the development work of the RPV. In view of the vehicle characteristics, a new systemic method called system-theoretic process analysis (STPA) has been tentatively applied to the hazardous factor analysis of the RPV flight test. An uncontrolled system behavior "path sagging phenomenon" is identified by implementing a three degrees of freedom simulation based on wind tunnel test data and establishing landing safety system dynamics archetype. To obtain higher safety design effectiveness and considering safety design precedence, a longitudinal "belly-flap" control surface is innovatively introduced and designed to eliminate hazards in landing. Finally, flight tests show that the unsafe factor has been correctly identified and the landing safety has been efficiently improved.     

远程操控飞行器自适应神经网络观测器设计

针对无人飞行器远程操控系统设计时,由于远程操控飞行器动力学的非线性和飞行器控制系统性能的不确定性,无法精确建立远程操控飞行器控制系统模型的问题,提出了一种自适应神经网络状态观测器设计方法实现对远程操控飞行器的控制系统模型的估计。首先将飞行器的动力学环节与自动驾驶仪构成的闭环回路作为一个整体建立了远程操控飞行器控制系统的非线性模型。然后针对模型中存在未建模动态的问题,采用神经网络算法对非线性动力学模型进行在线辨识,并引入鲁棒项对附加扰动进行抑制。最后设计自适应律对神经网络的权值进行实时调整,保证了系统的稳定性,并基于Lyapunov理论证明了观测器的估计误差是最终一致有界的。仿真结果表明,所设计的观测器能够保证远程操控飞行器在存在未建模动态和附加扰动的情况下对飞行状态具有良好的估计性能。