基于视觉的UCAV自主着陆蒙特卡洛仿真研究

Monte Carlo Simulation for Vision-based Autonomous Landing of Unmanned Combat Aerial Vehicles

介绍了不确定性分析方法的应用以及不确定性的数学模型。回顾了蒙特卡洛仿真在飞控系统的鲁棒分析、鲁棒控制律设计以及灵敏度分析等方面的研究成果。设计并实现了基于MATLAB分布计算引擎(MATLAB Distributed Computing Engine)的分布式蒙特卡洛仿真方案,开发了可以创建分布式仿真任务的蒙特卡洛仿真工具MCST(Monte Carlo Simulation Tool),可以方便地对于Simulink模型进行蒙特卡洛仿真。并且使用该工具进行基于视觉的UCAV自主着陆仿真研究。

Application of uncertainty analysis methods was introduced as well as some general used uncertainty models. The most basic uncertainty analysis method is Monte Carlo Simulation,which is a powerful and practical tool to deal with non-linear systems with very large amount of uncertainties. However,the great disadvantage of Monte Carlo is that it is very intensive computationally. Given to such a drawback,a distributed computing tool,which is named Monte Carlo Simulation Tool and based on MATLAB Distributed Computing Engine and Distributed Computing Toolbox,was developed in order to execute independent MATLAB operations simultaneously on a cluster of computers,speeding up execution of large amount of simulations. The MCST can easily be used to Simulink models for Monte Carlo Simulation. Monte Carlo Simulation has been applied to vision-based autonomous landing of Unmanned Combat Aerial Vehicles （UCAVs） with uncertainties of initial conditions and sensor measurements. Simulation results show that there is a high mission successful probability,which means the autonomous landing control law is insensitive to uncertainties.