基于IAPF算法和Arduino单片机的桥检无人机研究

发布时间：2018-12-31 21:48

【摘要】：无人机(UAV)技术可使桥梁检测的效率大为提高,且在经济性方面具有较强的优越性。但目前的无人机仍无法满足桥梁检测的工程条件和要求,主要有以下两方面的问题:1)无人机无避障功能,桥梁检测过程中存在诸多的安全隐患;2)无人机图传系统未能与桥梁检测相结合。针对这两个问题,本文从理论上和实践上提出切实可行的方法。主要从以下几方面展开。首先,阐述人工势场法在无人机进行桥梁检测避障的优越性。针对传统的人工势场法存在局部最小点和目标不可达的两个问题,根据桥梁结构自身的特点,提出改进人工势场法(IAPF)。并进行相应的仿真测试验证该算法的有效性。其次,阐述桥梁检测无人机硬件和软件的具体实现。引入Pixhawk飞控、Arduino单片机以及超声波传感器,在无避障功能的无人机基础上,添加了自动避障系统的硬件装置,并进行软件编写,并进行实际测试。再次,进行在裂缝检测的应用研究。引入图像传输技术和数字图像处理技术,通过图像信息计算出实际裂缝的长度和宽度,最后进行对照试验探究其效果。最后,试验验证改进后的无人机在桥梁检测中的可行性。在介绍该桥梁的情况之后,引入对照试验,将无人机检测的方法和常规检测方法引入该桥的外观检测,分别介绍了两种检测方法的实施方法和评分结果。对这两种方法分别从检测手段的图像效果、经济性和完成效率上进行比较。经测试,无人机进行检测具有较好地清晰度和可靠的评分结果,在经济性和效率上具有较大的优越性,其满足桥梁检测工程实践的要求,在桥梁检测应用中具有巨大潜力。
[Abstract]:UAV (UAV) technology can greatly improve the efficiency of bridge detection, and has strong economic advantages. However, the current UAV still can not meet the engineering conditions and requirements of bridge detection, there are two main problems: 1) the UAV has no obstacle avoidance function, there are many hidden dangers in the bridge detection process; 2) Unmanned aerial vehicle (UAV) map transmission system can not be combined with bridge detection. In view of these two problems, this paper puts forward practical methods in theory and practice. Mainly from the following aspects. Firstly, the superiority of artificial potential field method in bridge detection and obstacle avoidance of UAV is described. In view of the two problems of local minimum point and unreachable target in traditional artificial potential field method, according to the characteristics of bridge structure itself, an improved artificial potential field method (IAPF). Is proposed. The validity of the algorithm is verified by corresponding simulation tests. Secondly, the realization of the hardware and software of bridge detection UAV is expounded. With the introduction of Pixhawk flight control, Arduino microcontroller and ultrasonic sensor, the hardware device of the automatic obstacle avoidance system is added on the basis of the UAV without obstacle avoidance function, and the software is compiled and tested in practice. Thirdly, the application of crack detection is studied. Image transmission technology and digital image processing technology are introduced to calculate the actual crack length and width through image information. Finally, the feasibility of the improved UAV in bridge detection is verified. After introducing the situation of the bridge, the paper introduces the control test, introduces the UAV detection method and the conventional detection method into the bridge appearance inspection, and introduces the implementation method and the score result of the two detection methods respectively. The two methods are compared in image effect, economy and completion efficiency. The test results show that the UAV has good clarity and reliable scoring results and has great advantages in economy and efficiency. It meets the requirements of bridge detection engineering practice and has great potential in bridge detection application.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U446

【参考文献】