深水长基线定位导航技术研究

发布时间：2018-04-28 11:34

本文选题：长基线定位 + 波形设计　；参考：《哈尔滨工程大学》2014年博士论文

【摘要】：随着海洋技术的飞速发展,深海矿产资源的开发受到越来越多的关注,AUV作为一种便捷的勘探工具被广泛应用。而长基线定位技术是保障深海环境中AUV安全作业和高精度导航的必备手段之一。论文以深海大范围勘探作业对长基线定位的需求为背景,在分析比对了不同误差源对定位精度影响程度的基础上,研究了适用于低速目标的多标序列生成算法与检测技术、基阵网的整体平差与标定技术以及基于卡尔曼滤波的运动补偿技术。对于长基线定位系统而言,稳定可靠的测距技术是高精度定位的前提,具有优良检测估计性能的跳频编码序列为此提供了便利途径。但深海矿产资源的勘探往往具有探测覆盖面积广的特点,为此需要投放大量的声信标以满足区域内的高精度定位服务,以Costas阵列为代表的构造方法为保证序列的相关抑制能力,普遍存在着序列数量与信号频带宽度、脉冲长度相制约的缺陷。论文以Costas阵列追求完美的图钉式模糊函数的设计思想为引导,提出了以牺牲信号在大多普勒频偏条件下的相关性能为代价的序列生成方法,以解决多用户的识别问题。针对典型的检测干扰,研究了基于拷贝相关波形的压缩特性、脉冲对组合检测以及拷贝相关检测与能量检测联合检测的改进方法。仿真与试验证明,方法在保证了高精度测距的同时,有效的解决了干扰问题。多信标阵型扩大了长基线覆盖范围,但同时也引入了标定精度和标定效率的问题。基于测量船航行的标定方法受深海声速修正和船摇摆补偿精度不够的影响,标定精度很难提高,从而导致定位精度下降;另外,大量信标的标定时间甚至超出了实际勘探时间,严重的影响了作业效率。为此,论文提出了一种基于基阵网整体平差的联合标定方法,利用高精度的测距信息,提高了阵型的相对精度和标定效率。仿真和试验证明,方法对相对阵型标定精度有明显的改善效果。依靠声信号检测的长基线定位系统,不可避免的存在定位数据无效或者输出错误数据的现象而且数据输出速率偏低,这对于实时性要求较高的自主导航系统而言是难以接受的。卡尔曼滤波技术对于数据稳定性和输出速率的提升具有良好的效果,但由于长基线声学测距信息更新慢,外界干扰严重,致使常规卡尔曼滤波模型极易发散。本文提出了基于时延与位置互反馈的发散抑制方法,克服了估计模型与实际运动状态不匹配的问题。利用测距采样时刻补偿的方法改善了目标运动引入的定位误差,并对附有深度信息融合的解算模型进行了分析,给出了一种基于深度解算的参数误差检测方法。试验证明,改进的长基线定位解算方法提高了系统的稳定性和可靠性。
[Abstract]:With the rapid development of marine technology, more and more attention has been paid to the exploitation of deep-sea mineral resources. AUV is widely used as a convenient exploration tool. Long baseline positioning technology is one of the necessary means to ensure AUV safe operation and high precision navigation in deep sea environment. In this paper, based on the requirement of long baseline positioning in deep sea exploration, based on the analysis and comparison of the influence of different error sources on positioning accuracy, the algorithm and detection technology of multi-scale sequence generation suitable for low speed targets are studied. The integral adjustment and calibration technology of array network and the motion compensation technology based on Kalman filter. For long baseline positioning system, stable and reliable ranging technology is the premise of high precision location, and frequency hopping coding sequence with excellent detection and estimation performance provides a convenient way for this purpose. However, the exploration of deep-sea mineral resources often has the characteristics of wide exploration coverage, so a large number of acoustic beacons need to be put in to meet the high-precision positioning services in the area, and the structural method represented by Costas array is used to ensure the relevant suppression ability of the sequence. The number of sequences is limited by the frequency band width and pulse length. Guided by the design idea of the perfect map nail ambiguity function of Costas array, this paper presents a series generation method based on the correlation of sacrificial signals under the condition of large Doppler frequency offset to solve the problem of multi-user identification. For typical detection interference, the compression characteristics based on copy correlation waveform, pulse pair combination detection and improved method of joint detection of copy correlation detection and energy detection are studied. Simulation and experiments show that the method not only ensures high precision ranging, but also effectively solves the interference problem. Multi-beacon array expands the coverage of long baseline, but also introduces the problems of calibration accuracy and calibration efficiency. The calibration method based on the navigation of the survey ship is affected by the deep sea sound velocity correction and the ship swaying compensation accuracy is not enough, so it is difficult to improve the calibration accuracy, which leads to the decrease of the positioning accuracy. In addition, the calibration time of a large number of beacons is even longer than the actual exploration time. A serious impact on the efficiency of the operation. For this reason, this paper proposes a joint calibration method based on the integral adjustment of the array network. The relative precision and calibration efficiency of the array are improved by using the high-precision ranging information. Simulation and experiments show that the accuracy of relative matrix calibration is improved obviously. In the long baseline positioning system which depends on acoustic signal detection, it is inevitable that the localization data is invalid or the output error data rate is low, which is difficult to accept for the autonomous navigation system with high real-time requirements. The Kalman filtering technique has good effect on improving the data stability and the output rate, but the conventional Kalman filter model is easy to divergence due to the slow updating of the long baseline acoustic ranging information and the serious external interference. In this paper, a divergence suppression method based on time delay and position feedback is proposed, which overcomes the mismatch between the estimation model and the actual motion state. The location error introduced by target motion is improved by the method of distance measurement sampling time compensation, and the solution model with depth information fusion is analyzed, and a parameter error detection method based on depth solution is presented. Experimental results show that the improved long baseline positioning method improves the stability and reliability of the system.
【学位授予单位】：哈尔滨工程大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：P744;U666

【相似文献】