基于数据融合的井下人员定位方法研究与实现

发布时间：2019-03-15 09:07

【摘要】：社会的发展依赖能源的支撑,煤矿是我国三大能源之一。随着工业的发展,工业对煤矿的需求量日益增加,给煤矿工业的发展提出了更大的挑战。然而近年来煤矿井下的瓦斯泄露、透水、坍塌和煤尘爆炸等事故频繁发生,给煤矿生产带来了巨大的灾难和损失,煤矿开采的安全成为了煤炭行业和整个社会关注的焦点。煤矿井下人员定位系统的出现在一定程度上保障了开采人员的安全,它在灾害预警、逃生指导、人员搜救中发挥着重要的作用。定位精度和稳定性是井下人员定位系统的两大主要性能指标。而现阶段的人员定位系统都存在精度差、稳定性差的缺点。为了提高定位系统的精度,需要选择价格昂贵、精度高的传感器,这会增加定位系统的成本,不利于其普及和推广。随着信息融合技术的发展,将多种定位算法进行融合,互相取长补短,将为人员定位系统精度的提高提供新的解决思路和方法。论文选用JN5148模块,设计完成了带TOF和RSSI测距的Zigbee模块,利用该模块搭建了井下人员定位系统的实验平台。平台分为井上和井下部分,井上部分为总线通信,井下部分为Zigbee无线通信。论文主要对井下部分的分站、读卡器、标示卡之间的定位算法进行研究。首先论文在Eclipse软件开发平台上,实现了Zigbee组网;通过实时采集节点间RSSI和TOF测距数据,对不同环境下的不同干扰因素进行了分析和处理,在Matlab仿真平台上研究实现了测距的均值滤波、Kalman滤波和加权平均滤波等方法;在实现稳定测距的基础上,进一步研究了三边定位、Bounding-Box定位算法,提出了加权平均的测距融合算法和Bounding-Box结合Kalman滤波的二维定位融合算法,分别在仿真平台和实验平台上验证了该算法;并在实际煤矿环境下,分别对测距和定位算法现场调试。论文在井下人员定位系统硬件测试平台上对论文提出的信息融合算法进行了验证测试。结果表明论文提出的算法能够提高定位精度和人员定位系统的稳定性,达到了预期研究目标。
[Abstract]:The development of society depends on the support of energy. Coal mine is one of the three major energy sources in China. With the development of industry, the demand for coal mine is increasing day by day, which poses a greater challenge to the development of coal mine industry. However, in recent years, accidents such as gas leakage, water permeability, collapse and coal dust explosion occurred frequently in coal mines, which brought huge disasters and losses to coal mine production. The safety of coal mining has become the focus of attention in the coal industry and the whole society. Coal mine personnel positioning system to a certain extent to ensure the safety of mining personnel, it plays an important role in disaster early warning, escape guidance, personnel search and rescue. Positioning accuracy and stability are two main performance indexes of underground personnel positioning system. At present, the personnel positioning system has the shortcomings of poor precision and poor stability. In order to improve the accuracy of the positioning system, it is necessary to select expensive and high-precision sensors, which will increase the cost of the positioning system, which is not conducive to its popularization and popularization. With the development of information fusion technology, many location algorithms will be fused to complement each other, which will provide a new solution to improve the accuracy of personnel positioning system. In this paper, the JN5148 module is selected to design and complete the Zigbee module with TOF and RSSI ranging, and the experimental platform of underground personnel positioning system is built by using this module. The platform is divided into underground part and downhole part. The downhole part is bus communication and the downhole part is Zigbee wireless communication. This paper mainly studies the location algorithm between the sub-station, card reader and indicator card in the underground part. First of all, on the Eclipse software development platform, the paper realizes the Zigbee networking; Through real-time collection of RSSI and TOF ranging data between nodes, different interference factors in different environments are analyzed and processed. The mean filtering, Kalman filtering and weighted average filtering of ranging are studied and implemented on the Matlab simulation platform. Based on the realization of stable ranging, the trilateral location and Bounding-Box localization algorithm are further studied, and the weighted average ranging fusion algorithm and the two-dimensional location fusion algorithm based on Bounding-Box and Kalman filtering are proposed. The algorithm is verified on the simulation platform and the experimental platform. And in the actual coal mine environment, the ranging and positioning algorithm is debugged on the spot. In this paper, the information fusion algorithm is tested on the hardware test platform of underground personnel positioning system. The results show that the proposed algorithm can improve the positioning accuracy and the stability of the personnel positioning system, and achieve the expected research goal.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD655;TP202

【相似文献】