【作者】 程翔；

【导师】 熊凌；

【作者基本信息】 武汉科技大学 ， 控制科学与工程， 2018， 硕士

【摘要】 水下目标探测是人类认识与了解海洋的重要途径,探测的精度一直是热点问题。水下潜器是进行目标探测的重要工具,而保证水下潜器自身安全是完成任务的必要前提。由于地质的变化、基准图的精度有限、以及基准图年代久远等原因,对前方的环境不了解造成与不明物相碰撞是威胁水下潜器安全的原因之一;其次由于目前大部分水下目标探测技术在进行目标探测时自身会发射信号,会破坏自身的隐蔽性,因此一种无源的基于重力场水下目标探测方法十分有意义。基于重力场的水下目标探测方法以及探测误差一直是水下目标探测的研究重点,本文首先从重力场的基本原理出发,分析出基于重力场目标探测的可行性。通过仿真数据,得出了仅仅密度不相同的情况下,重力梯度值与密度成正比的关系。其次,将仿真得到的海水、岩石、以及潜艇模型的重力梯度数据描绘出来,分析出当重力梯度值精度为10^-5E时,探测的范围为1000m,故后文也由此来设定航线进行基于重力场的水下目标探测。然后,本文进行了基于重力与重力梯度融合目标探测,将该方法与仅使用重力梯度的目标探测方法对比。探测的精度一直是水下目标探测的重要评价指标之一,故本文利用数值分析的方法进行探测误差分析,得到重力与重力梯度融合目标探测的误差在5%以内。并进行仿真验证,仿真证明将重力与重力梯度融合进行水下目标探测的效果优于仅使用重力梯度的目标探测方法,其误差在允许范围之内,可以为水下潜器完成水下目标探测的任务。最后,本文还提出基于牛顿迭代法的水下重力梯度目标探测方法,将目标探测的相对位置转化成三维坐标系下三个非线性方程组成的非线性方程组求解问题。仿真实验结果表明,基于牛顿迭代法的水下重力梯度目标探测能够较好的实现水下目标探测。相比其他方法,该方法仅仅使用重力梯度仪就能将探测精度比重力与重力梯度融合目标探测的精度高1%,精度得到了提高。该目标探测仅仅使用重力梯度仪,相比基于重力与重力梯度目标探测,减少了成本。更多还原

【Abstract】 Underwater target detection is an important way for humans to understand and understand the ocean.The accuracy of detection has always been a hot issue.The underwater vehicle is an important tool for target detection,and ensuring the safety of the underwater vehicle itself is a necessary prerequisite for the completion of the mission.Due to changes in geology,limited accuracy of reference maps,and long-standing reference maps,it is one of the reasons that threaten the safety of underwater vehicles due to the lack of understanding of the front environment and the collision with unknown objects.Secondly,because most underwater target detection technologies transmit their own signals during target detection,which will undermine their concealment,a passive underwater-based target detection method based on gravitational field is very meaningful.The underwater target detection method and detection error based on the gravity field have always been the focus of underwater target detection.This paper firstly analyzes the feasibility of the target detection based on the gravity field from the basic principle of the gravity field.Through the simulation data,it is found that the gravity gradient value is directly proportional to the density when only the density is different.Secondly,the gravity gradient data of seawater,rocks,and submarine models obtained from the simulation are plotted.It is analyzed that when the accuracy of the gravity gradient value is 10-5E,the detection range is 1000 m.Therefore,the following text is used to set the route to carry out.Underwater target detection based on gravity field.Then,fusion detection based on gravity and gravity gradient is performed in this paper,and the method is compared with the target detection method using only gravity gradient.The accuracy of detection has always been one of the important evaluation indicators for underwater target detection.Therefore,this paper uses the method of numerical analysis to analyze the detection error,and the error of fusion of gravity and gravity gradient target detection is within 5%.Simulation and verification are carried out.The simulation proves that the effect of fusion of gravity and gravity gradients for underwater target detection is better than that using only the gravity gradient.The error is within the allowable range and underwater target can be detected for underwaterFinally,this paper also proposes a underwater gravity gradient target detection method based on Newton iteration method,which transforms the relative position of target detection into a nonlinear equation set composed of three nonlinear equations in a three-dimensional coordinate system.The simulation experiment results show that the underwater gravity gradient target detection based on Newton iteration method can achieve the underwater target detection better.Compared with other methods,this method can only use the gravity gradiometer to compare the accuracy of the detection with gravity and gravity gradient fusion target detection accuracy of about 1%,and accuracy has been improved.This target detection uses only a gravity gradiometer,which reduces costs compared to target detection based on gravity and gravity gradients.更多还原