基于非线性滤波的水下地形辅助导航方法

为了解决水下航行器惯性导航误差随时间积累问题,利用地形辅助导航技术进行导航位置修正。由于水下地形的非线性,对基于扩展卡尔曼滤波(EKF)、无迹卡尔曼滤波(UKF)和粒子滤波(PF)3种非线性滤波方法的水下地形辅助导航算法进行研究。针对传统基于单波束声纳量测模型在小起伏地形区域导航精度低、易发散问题,从提高量测地形信息量角度,建立了基于多波束测深声纳的量测模型。使用多波束实测地形数据进行仿真试验,结果表明:无论在粗糙地形区域还是较平坦地形区域,多波束量测模型有效缓解了3种方法易发散问题,提高了导航精度。     

基于地球系统模式的局地化粒子滤波器与集合卡尔曼滤波器同化实验

粒子滤波器(PF)是一种非常具有应用前景的非线性资料同化方法。但由于其算法本身存在的粒子退化问题,目前尚未被广泛地应用于大型地球物理模式。目前主流的集合同化系统仍然倾向于使用集合卡尔曼滤波器(EnKF)及其衍生方法。一种新近被提出的局地化粒子滤波器(LPF)在经典的粒子滤波器算法中引入局地化技术,可以使用较小的计算成本有效地避免退化问题,具有非常大的业务应用潜力。本文在全耦合的通用地球系统模式中开展了LPF和EnKF的同化实验,同化资料为模拟的卫星海表温度资料。着重考察了不同局地化参数对两种方法的不同影响,对比了局地化粒子滤波器与集合卡尔曼滤波器的同化效果差异。比较的结果表明,LPF的同化效果对于局地化参数的选择非常敏感,在使用最优局地化参数的条件下,LPF能达到与EnKF相当甚至优于后者的同化效果,并具有较大的改进空间。     

几种非线性滤波算法的性能比较与分析

在非线性状态估计中,传统的扩展卡尔曼滤波通过线性化来实现高斯近似,由于截断误差的存在很难保证估计精度;而基本粒子滤波容易出现粒子退化,导致滤波发散。针对粒子滤波的两个基本假设：蒙特卡罗假设和重要采样假设,采用蒙特卡罗随机链的方法来提高粒子的多样性,并利用无味卡尔曼滤波来产生更高精度的替代分布,发展了无味粒子滤波。通过仿真实验证明,相比较扩展卡尔曼滤波和基本粒子滤波,改进后的无味粒子滤波算法性能更优越,对含有非线性非高斯的状态估计问题有更好的滤波效果。     

一种基于超短基线定位的便携式AUV回坞导航方法

自主水下机器人(AUV)对接技术是目前水下机器人的研究热点,精确可靠的AUV的回坞导航是实现对接的关键技术。对于追求轻便的便携式AUV的对接系统,考虑到便携式AUV的搭载能力有限又需要足够的定位精度用于对接,提出了一种基于超短基线(USBL)定位的回坞导航方法,该方法让AUV只需装载电子罗盘和水声应答器就能完成精确的回坞定位。根据导航方法的特点,设计了一种改进的扩展卡尔曼滤波算法,其优点是能在处理滞后的USBL数据的同时动态估算海流、更新状态方程以消除海流造成的定位误差。通过湖试和大量仿真实验,验证了定位算法在海流影响下的定位性能。     

Robust Range-Only Beacon Localization

In this paper, we present a system capable of simultaneously estimating the position of an autonomous underwater vehicle (AUV) and the positions of stationary range-only beacons. Notably, our system does not require beacon positions a priori, and our system performs well even when range measurements are severely degraded by noise and outliers. We present a powerful outlier rejection method that can identify groups of range measurements that are consistent with each other, and a method for initializing beacon positions in an extended Kalman filter (EKF). We have successfully applied our algorithms to real-world data and have demonstrated a simultaneous localization and mapping (SLAM) system whose navigation performance is comparable to that of systems that assume known beacon locations     

EnKF和SIR-PF在贝叶斯滤波框架下的比较和结合

贝叶斯估计理论为非线性、非高斯系统的数据同化提供了一个统一的框架。在本文中,我们利用著名的洛伦茨吸引子(Lorenz'63)模式对两种基于贝叶斯滤波理论的数据同化方法——集合卡尔曼滤波器(EnKF)和重取样粒子滤波器(SIR-PF)——进行了较为全面的比较。比较的结果揭示了两种方法的优缺点:即当集合成员数目较多时,SIR-PF的同化效果优于EnKF;反之,则EnKF的表现较好。进一步地,我们使用统计方法分析了两者表现的差异和原因。最近提出的一种集合卡尔曼粒子滤波器(EnKPF)通过使用一个可控的参数整合EnKF和SIR-PF的分析格式,可以结合两者的优点。本文在充分比较两种方法的前提下,重新阐释并改进了原有的EnKPF算法,使之适用于非线性的观测算子。通过使用相同的洛伦茨模式实验,我们揭示了EnKPF实质上提供了关于EnKF和SIR-PF的连续插值,使得后两者可以视为其特殊情况。并且,在集合成员数目有限的前提下,EnKPF可以在一定程度上避免滤波退化的发生,取得优于EnKF和SIR-PF的同化效果。     

Deep Sea AUV Navigation Using Multiple Acoustic Beacons

Navigation is a critical requirement for the operation of Autonomous Underwater Vehicles (AUVs). To estimate the vehicle position, we present an algorithm using an extended Kalman filter (EKF) to integrate dead-reckoning position with acoustic ranges from multiple beacons pre-deployed in the operating environment. Owing to high latency, variable sound speed multipath transmissions and unreliability in acoustic measurements, outlier recognition techniques are proposed as well. The navigation algorithm has been tested by the recorded data of deep sea AUV during field operations in a variety of environments. Our results show the improved performance over prior techniques based on position computation.     

基于海气耦合模式的均权重粒子滤波与集合卡尔曼滤波比较研究

在大气和海洋环境研究中,粒子滤波(PF)由于在非线性数据同化方面突出的优势,逐渐成为研究热点。最近改进的均权重粒子滤波(EWPF)为粒子滤波的进一步发展指明了新方向。集合卡尔曼滤波方法 (EAKF)作为当前主要应用的数据同化方法,使用高斯假设和线性假设来解决非线性问题,然而对均权重粒子滤波方法和卡尔曼滤波方法在非线性模式下的同化结果和特点还缺少系统详细的比较研究。本文在非线性耦合气候模式下,比较研究两种同化方法,采用均方根误差(RMSE)作为评价比较标准。实验结果表明,在非线性低频观测耦合模式中EWPF结果均优于EAKF。同时根据RMSE的结果得出,EWPF的同化结果更接近观察结果,而EAKF的同化结果更接近模式真值。     

基于CT非线性模型的水下目标跟踪算法比较

针对水下目标跟踪非线性跟踪精度问题,假设目标机动模型为恒转速运动模型,贝叶斯框架下,因扩展卡尔曼滤波跟踪方法进行模型在估计点的泰勒展开,忽略一阶以上高阶项,存在模型误差,比较了扩展卡尔曼滤波、无迹卡尔曼滤波、容积卡尔曼滤波在高斯噪声干扰下滤波误差均方根,以及3种方法运行时间。仿真证明,非线性系统下状态维度为5,容积卡尔曼滤波跟踪的精度高于无迹卡尔曼滤波,无迹卡尔曼滤波高于扩展卡尔曼滤波。该研究为海上目标非线性测量系统提供仿真实例,为进一步滤波算法改进提供基础。     

多传感器组合导航系统研究

为了满足水下航行器高精度导航定位的需求,建立了多传感器组合导航的系统模型。针对信息融合过程中出现的非线性环节,在传统联邦滤波器的基础上,提出了基于粒子滤波的混合联邦滤波器。其中,线性子系统采用卡尔曼滤波算法进行滤波估计,非线性子系统采用粒子滤波算法进行滤波估计。计算机仿真分析表明,该混合联邦滤波算法能够将线性和非线性子系统的滤波结果很好地融合起来,提高了组合导航系统的定位精度。     

集合资料同化方法的理论框架及其在海洋资料同化的研究展望

在海洋动力系统的数值模拟中,海洋资料同化是一种能够有效融合多源海洋观测资料和数值模式的方法。它不仅可以显著地提高数值模拟的效果,构造海洋再分析资料场,还能有效减少海洋和气候预报时模式初始条件的不确定性。因此,海洋资料同化对于海洋研究和业务化应用具有非常重要的意义。资料同化方法的研究一直是大气、海洋科学的热门课题之一。其中,集合卡尔曼滤波器(EnKF)是一种有效的资料同化方法,自提出以来经过了20多年的发展和改进,已经在海洋资料同化中得到了广泛的研究和应用。近年来,随着动力模式的不断发展和计算能力的提高,粒子滤波器由于不受模型线性和误差高斯分布假设的约束,也逐渐成为了当前资料同化方法研究的热点。本文分析和总结了目前关于集合卡尔曼滤波器和粒子滤波器的一些最新理论研究结果,在贝叶斯滤波理论的框架下讨论了这两类算法的关联和区别,以及各自在资料同化实践中的优势和不足。在此基础上,我们探讨了粒子滤波器应用于海洋模式资料同化的主要困难和目前可行的一些解决方法,展望了集合资料同化方法研究的新趋势,为集合资料同化方法的进一步发展和应用提供理论基础。     

数据融合下的短期潮位预报测试及分析

针对短期验潮数据分析难度大、预报精度低的问题,设计了一种基于数据融合技术的短期潮位预报方法,利用扩展卡尔曼滤波(EKF),将通用的潮汐模型计算水位融合到调和预报模型之中,生成精度更高的融合预报值.数据测试表明,对于3天的验潮数据,EKF方法至少对向后5天的预报有效,融合值较两种源数据的平均优化度分别为33％和60％;对...     

Simulation of an Inertial Acoustic Navigation System With Range Aiding for an Autonomous Underwater Vehicle

This paper presents an integrated navigation system for underwater vehicles to improve the performance of a conventional inertial acoustic navigation system by introducing range measurement. The integrated navigation system is based on a strapdown inertial navigation system (SDINS) accompanying range sensor, Doppler velocity log (DVL), magnetic compass, and depth sensor. Two measurement models of the range sensor are derived and augmented to the inertial acoustic navigation system, respectively. A multirate extended Kalman filter (EKF) is adopted to propagate the error covariance with the inertial sensors, where the filter updates the measurement errors and the error covariance and corrects the system states when the external measurements are available. This paper demonstrates the improvement on the robustness and convergence of the integrated navigation system with range aiding (RA). This paper used experimental data obtained from a rotating arm test with a fish model to simulate the navigational performance. Strong points of the navigation system are the elimination of initial position errors and the robustness on the dropout of acoustic signals. The convergence speed and conditions of the initial error removal are examined with Monte Carlo simulation. In addition, numerical simulations are conducted with the six-degrees-of-freedom (6-DOF) equations of motion of an autonomous underwater vehicle (AUV) in a boustrophedon survey mode to illustrate the effectiveness of the integrated navigation system.     

Parameter sensitivity of three Kalman filter schemes for assimilation of water levels in shelf sea models

In applications of data assimilation algorithms, a number of poorly known assimilation parameters usually need to be specified. Hence, the documented success of data assimilation methodologies must rely on a moderate sensitivity to these parameters. This contribution presents a parameter sensitivity study of three well known Kalman filter approaches for the assimilation of water levels in a three dimensional hydrodynamic modelling system. The filters considered are the ensemble Kalman filter (EnKF), the reduced rank square root Kalman filter (RRSQRT) and the steady Kalman filter. A sensitivity analysis of key parameters in the schemes is undertaken for a setup in an idealised bay. The sensitivity of the resulting root mean square error (RMSE) is shown to be low to moderate. Hence the schemes are robust within an acceptable range and their application even with misspecified parameters is to be encouraged in this perspective. However, the predicted uncertainty of the assimilation results are sensitive to the parameters and hence must be applied with care. The sensitivity study further demonstrates the effectiveness of the steady Kalman filter in the given system as well as the great impact of assimilating even very few measurements.     

水下自主机器人水动力预报方法研究

针对水下机器人操纵性优化设计中水动力系数预报问题,在水下机器人水动力预报中引入艇体肥瘦指数概念,确定了水下机器人艇体几何描述的五参数模型。提出采用小波神经网络方法预报水下机器人水动力,确定了神经网络的结构,利用均匀试验设计方法,设计了神经网络的学习样本。研究结果表明,只要确定适当的输入参数,选择适当的学习样本和网络结构,利用小波神经网络方法对水下机器人水动力进行预报可以达到较好的精度。     

The representer method, the ensemble Kalman filter and the ensemble Kalman smoother: A comparison study using a nonlinear reduced gravity ocean model

This paper compares contending advanced data assimilation algorithms using the same dynamical model and measurements. Assimilation experiments use the ensemble Kalman filter (EnKF), the ensemble Kalman smoother (EnKS) and the representer method involving a nonlinear model and synthetic measurements of a mesoscale eddy. Twin model experiments provide the “truth” and assimilated state. The difference between truth and assimilation state is a mispositioning of an eddy in the initial state affected by a temporal shift. The systems are constructed to represent the dynamics, error covariances and data density as similarly as possible, though because of the differing assumptions in the system derivations subtle differences do occur. The results reflect some of these differences in the tangent linear assumption made in the representer adjoint and the temporal covariance of the EnKF, which does not correct initial condition errors. These differences are assessed through the accuracy of each method as a function of measurement density. Results indicate that these methods are comparably accurate for sufficiently dense measurement networks; and each is able to correct the position of a purposefully misplaced mesoscale eddy. As measurement density is decreased, the EnKS and the representer method retain accuracy longer than the EnKF. While the representer method is more accurate than the sequential methods within the time period covered by the observations (particularly during the first part of the assimilation time), the representer method is less accurate during later times and during the forecast time period for sparse networks as the tangent linear assumption becomes less accurate. Furthermore, the representer method proves to be significantly more costly (2–4 times) than the EnKS and EnKF even with only a few outer iterations of the iterated indirect representer method.     

The statistical observation localized equivalent-weights particle filter in a simple nonlinear model

This paper presents an improved approach based on the equivalent-weights particle filter(EWPF) that uses the proposal density to effectively improve the traditional particle filter. The proposed approach uses historical data to calculate statistical observations instead of the future observations used in the EWPF’s proposal density and draws on the localization scheme used in the localized PF(LPF) to construct the localized EWPF. The new approach is called the statistical observation localized E...     

Stability Analysis on Speed Control System of Autonomous Underwater Vehicle

The stability of the motion control system is one of the decisive factors of the control quality for Autonomous Underwater Vehicle (AUV).The divergence of control,which the unstable system may be brought about,is fatal to the operation of AUV.The stability analysis of the PD and S-surface speed controllers based on the Lyapunov' s direct method is proposed in this paper.After decoupling the six degree-of-freedom (DOF) motions of the AUV,the axial dynamic behavior is discussed and the condition is deduced,in which the parameters selection within stability domain can guarantee the system asymptotically stable.The experimental results in a tank and on the sea have successfully verified the algorithm reliability,which can be served as a good reference for analyzing other AUV nonlinear control systems.     

Robust Nonlinear Path-Following Control of an AUV

This paper develops a robust nonlinear controller that asymptotically drives the dynamic model of an autonomous underwater vehicle (AUV) onto a predefined path at a constant forward speed. A kinematic controller is first derived, and extended to cope with vehicle dynamics by resorting to backstepping and Lyapunov-based techniques. Robustness to vehicle parameter uncertainty is addressed by incorporating a hybrid parameter adaptation scheme. The resulting nonlinear adaptive control system is formally shown and it yields asymptotic convergence of the vehicle to the path. Simulations illustrate the performance of the derived controller .      

A conservative nonlinear filter for the high-frequency range of wind wave spectra

Filtering of the high-frequency part of a wind wave spectrum may be useful in a numerical wind wave model for various reasons. First, it can be used to augment (or be part of) a parameterization of the resonant nonlinear interactions, that are essential to third-generation wind wave models. Second, when combined with a dynamic time stepping scheme for source term integration, it may result in smoother (and hence faster) wave model integration. In this study, such a filter is proposed, based on the traditional Discrete Interaction Approximation (DIA) for the resonant four-wave nonlinear interactions. This filter retains all conservative properties of the interactions. For small time steps and/or smooth spectra, it is formulated as a traditional source term. For larger time steps and/or non-smooth spectra it is formulated as a filter. This formulation guarantees stability of the filter itself and will enhance overall computational stability in a full wave model. The stability properties of this filter are illustrated using traditional wave growth computations. Examples are given where the filter improves model economy, and where it is shown to remove spurious high-frequency noise from a wave model.