无人机飞控系统的多源仿真研究

发布时间：2019-06-29 14:11

【摘要】：随着无人机技术的发展,一个无人机研发团队涉及到的专业越来越多,要完成无人机整体系统的仿真验证,需要各个学科之间完成各自负责的子系统模型的交互仿真,无人机飞控系统仿真也是如此。为了能够满足无人机飞控系统研发过程中需要完成大量多个专业工具生成的多源模型的联合仿真,在FMI标准的基础上,开展了无人机飞控系统多源仿真的相关技术研究。主要有多源模型输入数据外推算法,多源模型仿真误差估计和误差控制算法,以及多源模型仿真调度方法,并在此基础上开发了无人机飞控系统多源仿真平台和飞控系统仿真数据分析软件。在分析无人机飞控仿真基本理论的基础上,进一步分析了无人机飞控系统多源仿真的技术要求。从基于已有仿真模型接口开发多源仿真接口,利用HLA标准和利用FMI标准三种技术方向上确定了采用FMI标准的多源仿真方案。分析了多源模型仿真数据同步特点和多源模型并行化仿真的特点,在最小均方误差多项式外推方法的基础上提出了基于数据变化的多源仿真数据外推算法,使得分离模型完成并发运行具有可行性。但这种方式并不能保证并行化仿真的误差会被限制在某个范围内。为了保证仿真精度和仿真效率,在分析多源模型仿真结构的基础上,分析了多源模型由于数据外推造成系统全局误差的有界性。在理查逊外推法的基础上改进了多源模型并行仿真的误差估计方法,使得误差估计更加准确。并在更加准确的误差估计基础上,通过控制多源子系统模型之间的数据交互时间间隔,完成对误差范围的控制,达到仿真误差限定的效果。在并行仿真的基础上,研究了整个无人机飞控系统的仿真调度方法,将系统仿真调度分为了三级调度方式,分别是系统级别的并行化仿真方法,局部的串行化调度方法和单个模型的状态转移方法。最后,在分析了无人机飞控系统多源仿真需求的基础上,设计和开发了无人机飞控系统多源仿真平台和飞控仿真数据分析软件。支持蒙特卡罗随机参数化的仿真,和大量的仿真数据分析。最终为无人机飞控仿真提供了相对完整的解决方案。
[Abstract]:With the development of UAV technology, a UAV research and development team involves more and more specialties. In order to complete the simulation and verification of UAV overall system, it is necessary to complete the interactive simulation of their own responsible subsystem models among various disciplines, and the same is true of UAV flight control system simulation. In order to meet the need to complete the joint simulation of multi-source models generated by a large number of professional tools in the research and development of UAV flight control system, the related technology of multi-source simulation of UAV flight control system is studied on the basis of FMI standard. There are mainly multi-source model input data extrapolation algorithm, multi-source model simulation error estimation and error control algorithm, and multi-source model simulation scheduling method. On this basis, the multi-source simulation platform of UAV flight control system and the simulation data analysis software of flight control system are developed. Based on the analysis of the basic theory of UAV flight control simulation, the technical requirements of multi-source simulation of UAV flight control system are further analyzed. The multi-source simulation interface based on the existing simulation model interface is developed, and the multi-source simulation scheme based on FMI standard is determined by using HLA standard and FMI standard. The synchronization characteristics of multi-source model simulation data and the characteristics of multi-source model parallelization simulation are analyzed. Based on the least mean square error polynomial extrapolation method, a multi-source simulation data extrapolation algorithm based on data change is proposed, which makes the separation model feasible to complete concurrent operation. However, this method can not guarantee that the error of parallelization simulation will be limited to a certain range. In order to ensure the simulation accuracy and efficiency, based on the analysis of the simulation structure of the multi-source model, the bounded global error of the system caused by the data extrapolation of the multi-source model is analyzed. On the basis of Richarson extrapolation method, the error estimation method of multi-source model parallel simulation is improved, which makes the error estimation more accurate. On the basis of more accurate error estimation, the error range is controlled by controlling the data interaction time interval between the multi-source subsystem models, and the simulation error limit is achieved. On the basis of parallel simulation, the simulation scheduling method of the whole UAV flight control system is studied. The system simulation scheduling is divided into three levels: system level parallelization simulation method, local serial scheduling method and single model state transition method. Finally, based on the analysis of the multi-source simulation requirements of UAV flight control system, the multi-source simulation platform and flight control simulation data analysis software of UAV flight control system are designed and developed. Support Monte Carlo stochastic parametric simulation, and a large number of simulation data analysis. Finally, it provides a relatively complete solution for UAV flight control simulation.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：V279;V249.1

【相似文献】