水下机器人地磁辅助导航算法研究

发布时间：2018-06-28 11:30

本文选题：地磁导航 + 磁补偿　；参考：《哈尔滨工程大学》2016年博士论文

【摘要】：惯性导航系统(Inertial navigation system,INS)能够在全天候条件下,自主、隐蔽地进行连续三维空间定位和三维空间定向,为载体提供连续、实时的导航参数,在短时间内具有较高的导航精度。但是惯性导航系统的导航误差会随时间累积发散,在长时间导航过程中精度会降低,需要通过有效的辅助导航手段进行修正。地磁场是地球自身的固有场,而且是一个矢量场,地球空间上的任意一点都会有一个与其唯一对应的地磁矢量值。因此将地磁场作为辅助惯性导航系统的研究对象,具有理论意义和实际应用价值。现阶段,水下地磁辅助导航存在一些重要的技术难点如:水下机器人(Remote Operated Vehicle,ROV)磁干扰场的补偿、区域的地磁场模型、导航区域的地磁图适配性评价以及地磁匹配算法等。本论文针对上述问题展开研究,研究工作主要集中在以下几个方面:ROV载体磁干扰的补偿算法研究。根据载体产生的磁干扰场的特点以及在航行中的主要姿态特征,提出一种基于载体旋转测量的磁干扰补偿算法。传统的磁补偿方法主要是构建磁干扰场模型,求解模型参数。然而传统方法计算复杂,同时需要学习航行,在学习航行中会存在载体姿态的耦合现象,会导致参数求解容易陷入奇异,使得参数求解变得非常困难,补偿精度不容易控制。本文采用载体旋转代替载体的学习航行,简化模型的学习过程。在ROV载体旋转过程中得到磁干扰场与旋转角度的关系,建立对应的补偿模型。在航行过程中,利用该补偿模型对载体产生的磁干扰场进行实时补偿,提高补偿效率。区域地磁图构建及地磁图适配性评价算法研究。高精度地磁图是进行地磁辅助导航的前提条件。针对区域地磁图构建问题,在分析和总结现有区域地磁场模型的基础上,采用泰勒多项式地磁场模型构建了高精度区域地磁图,给出了模型的最优截止阶数和减小边缘效应的措施。地磁信息丰富的区域有利于INS导航误差的修正。针对区域地磁图适配性评价问题,本文基于多属性决策思想采用主成分分析法提取多个主要的地磁特征值,对区域适配性进行综合评价。该算法解决了采用单一特征值评价不全面的缺陷,且具有求解过程简单、易于应用等优点。仿真结果表明:采用主成分分析法获得的地磁图适配性评价值与该区域的匹配概率有较好的一致性。地磁匹配算法研究。基于Hausdorff距离的MAGCOM算法在无旋转偏差时精度高,且抗干扰能力强。采用角度限差的ICCP算法对旋转偏差的处理十分有效,但计算费时。本文对两种算法进行组合优化,采用误差椭圆减少组合算法区域搜索的耗时,采用移位寄存器提高算法的实时性。此外,本文还研究了候选区域地磁图适配性对匹配算法修正精度的影响。仿真结果表明:适配性评价值高的区域有利于提高匹配算法的修正精度。松花湖水域试验研究。为验证以上理论研究的正确性与有效性,开展了基于ROV的水下地磁辅助导航试验。试验表明:载体旋转的磁补偿算法能够有效地补偿载体产生的磁干扰场;采用主成分分析法评价的结果能够选出适合导航的区域;在合适的区域内,组合算法对惯性导航系统的累积误差具有很高的修正精度。
[Abstract]:Inertial navigation system (INS) can carry out continuous three-dimensional space positioning and three-dimensional spatial orientation independently and covertly in all weather conditions, provide continuous and real-time navigation parameters for the carrier, and have high navigation accuracy in a short time, but the navigation error of the inertial navigation system will accumulate with time. In the course of long time navigation, the precision will be reduced, and it needs to be corrected by the effective auxiliary navigation means. The geomagnetic field is the intrinsic field of the earth itself, and it is a vector field, and any point in the earth's space will have a magnetic vector value corresponding to it only. Therefore, the study of the geomagnetic field as an auxiliary inertial navigation system is a study of the geomagnetic field. It has theoretical significance and practical application value. At present, there are some important technical difficulties in underwater geomagnetic auxiliary navigation, such as the compensation of the Remote Operated Vehicle (ROV), the regional geomagnetic field model, the suitability evaluation of the geomagnetic map in the navigation area and the geomagnetic matching algorithm. The research work is mainly focused on the following aspects: the study of the compensation algorithm for ROV carrier magnetic interference. Based on the characteristics of the magnetic interference field produced by the carrier and the main attitude characteristics of the navigation, a magnetic interference compensation algorithm based on the vector rotation measurement is proposed. The main method of the transmission of the magnetic interference is to construct the magnetic interference field mode. However, the traditional method is complicated and needs to learn navigation. There will be a coupling phenomenon of carrier attitude in learning and navigation, which will lead to the parameter solution easy to fall into the singularity, making the parameter solution very difficult and the compensation precision is not easy to control. This paper uses the carrier rotation instead of the carrier to learn navigation, simplification The relationship between the magnetic interference field and the rotation angle is obtained in the ROV carrier rotation process, and the corresponding compensation model is established. In the course of navigation, the compensation model is used to compensate the magnetic interference field produced by the carrier in real time and improve the compensation efficiency. The high precision of the regional geomagnetic map construction and the geomagnetic map adaptation evaluation algorithm. Geomagnetic map is the precondition for geomagnetic auxiliary navigation. Aiming at the problem of regional geomagnetic map construction, based on the analysis and summary of the existing regional geomagnetic field model, the high precision regional geomagnetic map is constructed by using the Taylor polynomial geomagnetic field model. The optimal cut-off order and the measures to reduce the edge effect are given. The geomagnetic information is rich. The region is beneficial to the correction of INS navigation error. In view of the suitability evaluation of regional geomagnetic map, this paper uses the principle component analysis method to extract a number of major geomagnetic features based on the multi attribute decision thinking method, and makes a comprehensive evaluation on the regional suitability. The algorithm solves the problem of incomplete evaluation by using a single eigenvalue and has solved the problem. The simulation results show that the matching probability of the geomagnetic map adaption obtained by the principal component analysis method is in good agreement with the matching probability of the region. The geomagnetic matching algorithm is studied. The MAGCOM algorithm based on Hausdorff distance has high precision and strong anti-interference ability when it has no rotation deviation, and the angle difference ICCP is used. The algorithm is very effective for the rotation deviation processing, but the calculation is time-consuming. In this paper, the two algorithms are combined and optimized, the error ellipse is used to reduce the time consuming of the region search, and the shift register is used to improve the real-time performance of the algorithm. In addition, this paper also studies the influence of the candidate region geomagnetic map adaptation on the correction accuracy of the matching algorithm. The real results show that the region with high suitability evaluation is helpful to improve the correction accuracy of the matching algorithm. In order to verify the correctness and effectiveness of the above theoretical research, a ROV based underwater geomagnetic auxiliary navigation test is carried out to verify the correctness and effectiveness of the above theoretical research. The experiment shows that the magnetic compensation algorithm of the carrier rotation can effectively compensate the carrier generation. The magnetic interference field, which is evaluated by the principal component analysis method, can select the area suitable for navigation. In the appropriate area, the combined algorithm has a high correction accuracy for the cumulative error of the inertial navigation 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】：2016
【分类号】：TP242;TN96

【相似文献】