• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文学位 >Reliability Assessment for Low-cost Unmanned Aerial Vehicles.
【24h】

Reliability Assessment for Low-cost Unmanned Aerial Vehicles.

机译：低成本无人机的可靠性评估。

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Existing low-cost unmanned aerospace systems are unreliable, and engineers must blend reliability analysis with fault-tolerant control in novel ways. This dissertation introduces the University of Minnesota unmanned aerial vehicle flight research platform, a comprehensive simulation and flight test facility for reliability and fault-tolerance research. An industry-standard reliability assessment technique, the failure modes and effects analysis, is performed for an unmanned aircraft. Particular attention is afforded to the control surface and servo-actuation subsystem. Maintaining effector health is essential for safe flight; failures may lead to loss of control incidents. Failure likelihood, severity, and risk are qualitatively assessed for several effector failure modes. Design changes are recommended to improve aircraft reliability based on this analysis. Most notably, the control surfaces are split, providing independent actuation and dual-redundancy. The simulation models for control surface aerodynamic effects are updated to reflect the split surfaces using a first-principles geometric analysis.;The failure modes and effects analysis is extended by using a high-fidelity nonlinear aircraft simulation. A trim state discovery is performed to identify the achievable steady, wings-level flight envelope of the healthy and damaged vehicle. Tolerance of elevator actuator failures is studied using familiar tools from linear systems analysis. This analysis reveals significant inherent performance limitations for candidate adaptive/reconfigurable control algorithms used for the vehicle. Moreover, it demonstrates how these tools can be applied in a design feedback loop to make safety-critical unmanned systems more reliable.;Control surface impairments that do occur must be quickly and accurately detected. This dissertation also considers fault detection and identification for an unmanned aerial vehicle using model-based and model-free approaches and applies those algorithms to experimental faulted and unfaulted flight test data. Flight tests are conducted with actuator faults that affect the plant input and sensor faults that affect the vehicle state measurements. A model-based detection strategy is designed and uses robust linear filtering methods to reject exogenous disturbances, e.g. wind, while providing robustness to model variation. A data-driven algorithm is developed to operate exclusively on raw flight test data without physical model knowledge. The fault detection and identification performance of these complementary but different methods is compared. Together, enhanced reliability assessment and multi-pronged fault detection and identification techniques can help to bring about the next generation of reliable low-cost unmanned aircraft.
机译：现有的低成本无人机系统不可靠，工程师必须以新颖的方式将可靠性分析与容错控制相结合。本文介绍了明尼苏达大学的无人机飞行研究平台，这是一个用于可靠性和容错研究的综合模拟和飞行测试设备。对无人飞机执行了行业标准的可靠性评估技术，故障模式和影响分析。特别要注意控制面和伺服驱动子系统。保持效应器健康对于安全飞行至关重要。故障可能会导致失控事件。定性评估了几种效应器失效模式的失效可能性，严重性和风险。建议根据此分析进行设计更改以提高飞机的可靠性。最值得注意的是，控制表面是分开的，提供独立的致动和双重冗余。使用第一性原理几何分析更新了控制面空气动力学效果的仿真模型，以反映分裂的表面。通过使用高保真非线性飞机仿真扩展了失效模式和效果分析。进行修剪状态发现以识别健康且受损的车辆可达到的稳定的机翼水平飞行包线。使用线性系统分析中熟悉的工具来研究电梯执行器故障的容限。该分析揭示了用于车辆的候选自适应/可重构控制算法的显着固有性能限制。此外，它演示了如何将这些工具应用到设计反馈回路中，从而使对安全至关重要的无人驾驶系统更加可靠。必须迅速而准确地检测出确实发生的控制表面损伤。本文还考虑了基于模型和无模型的方法对无人机的故障检测和识别，并将这些算法应用于实验性故障和非故障飞行测试数据。使用影响工厂输入的执行器故障和影响车辆状态测量的传感器故障进行飞行测试。设计了基于模型的检测策略，并使用鲁棒的线性滤波方法来拒绝外部干扰，例如同时为模型变化提供鲁棒性。开发了一种数据驱动算法，以在没有物理模型知识的情况下仅对原始飞行测试数据进行操作。比较了这些互补但不同的方法的故障检测和识别性能。综合起来，增强的可靠性评估和多管齐下的故障检测与识别技术可以帮助实现下一代可靠的低成本无人飞机。

著录项

  • 作者

    Freeman, Paul Michael.;

  • 作者单位

    University of Minnesota.;

  • 授予单位 University of Minnesota.;
  • 学科 Aerospace engineering.
  • 学位 Ph.D.
  • 年度 2014
  • 页码 122 p.
  • 总页数 122
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号