Description: In this demo, I use the EM algorithm with a Rauch-Tung-Striebel smoother and an M step, which I ve recently derived, to train a two-layer perceptron, so as to classify medical data (kindly provided by Steve Roberts and Will Penny from EE, Imperial College). The data and simulations are described in: Nando de Freitas, Mahesan Niranjan and Andrew Gee Nonlinear State Space Estimation with Neural Networks and the EM algorithm After downloading the file, type \"tar -xf EMdemo.tar\" to uncompress it. This creates the directory EMdemo containing the required m files. Go to this directory, load matlab5 and type \"EMtremor\". The figures will then show you the simulation results, including ROC curves, likelihood plots, decision boundaries with error bars, etc. WARNING: Do make sure that you monitor the log-likelihood and check that it is increasing. Due to numerical errors, it might show glitches for some data sets.
Platform: |
Size: 198220 |
Author:晨间 |
Hits:
Description: 随着这些年计算机硬件水平的发展, 计算速度的提高, 源自序列蒙特卡罗方法的蒙特卡罗粒子滤波方法的应用研究又重新活跃起来。本文的这种蒙特卡罗粒子滤波算法是利用序列重要性采样的概念, 用一系列离散的带权重随机样本近似相
应的概率密度函数。由于粒子滤波方法没有像广义卡尔曼滤波方法那样对非线性系统做线性化的近似, 所以在非线性状态估计方面比广义卡尔曼滤波更有优势。在很多方面的应用已经逐渐有替代广义卡尔曼滤波的趋势。-With the years the level of computer hardware development, the speed of calculation, derived from the sequence of the Monte Carlo method, Monte Carlo particle filter method applied research has once again become active again. In this paper, this kind of Monte Carlo particle filter is to use the concept of sequence of the importance of sampling, using a series of discrete random sample with weights similar to the corresponding probability density function. Since the particle filtering method is not as broad as Kalman filtering method for nonlinear system to do linear approximation, nonlinear state estimation in the generalized Kalman filter than an advantage. Applications in many areas has been gradually generalized Kalman filter has an alternative trend. Platform: |
Size: 528384 |
Author:阳关 |
Hits:
Description: Carrier-phase synchronization can be approached in a
general manner by estimating the multiplicative distortion (MD) to which
a baseband received signal in an RF or coherent optical transmission
system is subjected. This paper presents a unified modeling and
estimation of the MD in finite-alphabet digital communication systems. A
simple form of MD is the camer phase exp GO) which has to be estimated
and compensated for in a coherent receiver. A more general case with
fading must, however, allow for amplitude as well as phase variations of
the MD.
We assume a state-variable model for the MD and generally obtain a
nonlinear estimation problem with additional randomly-varying system
parameters such as received signal power, frequency offset, and Doppler
spread. An extended Kalman filter is then applied as a near-optimal
solution to the adaptive MD and channel parameter estimation problem.
Examples are given to show the use and some advantages of this scheme. Platform: |
Size: 827392 |
Author:吴大亨 |
Hits:
Description: In this demo, I use the EM algorithm with a Rauch-Tung-Striebel smoother and an M step, which I ve recently derived, to train a two-layer perceptron, so as to classify medical data (kindly provided by Steve Roberts and Will Penny from EE, Imperial College). The data and simulations are described in: Nando de Freitas, Mahesan Niranjan and Andrew Gee Nonlinear State Space Estimation with Neural Networks and the EM algorithm After downloading the file, type "tar -xf EMdemo.tar" to uncompress it. This creates the directory EMdemo containing the required m files. Go to this directory, load matlab5 and type "EMtremor". The figures will then show you the simulation results, including ROC curves, likelihood plots, decision boundaries with error bars, etc. WARNING: Do make sure that you monitor the log-likelihood and check that it is increasing. Due to numerical errors, it might show glitches for some data sets.
-In this demo, I use the EM algorithm with a Rauch-Tung-Striebel smoother and an M step, which I ve recently derived, to train a two-layer perceptron, so as to classify medical data (kindly provided by Steve Roberts and Will Penny from EE, Imperial College). The data and simulations are described in: Nando de Freitas, Mahesan Niranjan and Andrew Gee Nonlinear State Space Estimation with Neural Networks and the EM algorithm After downloading the file, type "tar-xf EMdemo.tar" to uncompress it. This creates the directory EMdemo containing the required m files. Go to this directory, load matlab5 and type "EMtremor". The figures will then show you the simulation results, including ROC curves, likelihood plots, decision boundaries with error bars, etc. WARNING: Do make sure that you monitor the log-likelihood and check that it is increasing. Due to numerical errors, it might show glitches for some data sets.
Platform: |
Size: 197632 |
Author:晨间 |
Hits:
Description: Rebel工具包，包括UKF，CDKF在内的多种粒子滤波算法仿真，是非线性系统状态估计的良好帮手-Rebel kit, including the UKF, CDKF including a wide range of particle filter simulation, nonlinear system state estimation is a good helper Platform: |
Size: 1675264 |
Author:Eric |
Hits:
Description: 学习扩展卡尔曼滤波气的基本文件，可以随便下载并讨论-This is a tutorial on nonlinear extended Kalman filter (EKF). It uses the standard EKF fomulation to achieve nonlinear state estimation. Inside, it uses the complex step Jacobian to linearize the nonlinear dynamic system. The linearized matrices are then used in the Kalman filter calculation. Platform: |
Size: 55296 |
Author:tongliang |
Hits:
Description: 在机动目标跟踪中，机动目标模型是机动目标跟踪的基本要素之一，一般希望机动目标模型能准确表征目标机动时的各种运动状态。比较常用的模型有匀速运动(CV)模型、匀加速运动(CA) 模型、时间相关模型（Singer）和机动目标“当前”统计模型。上述模型均采用机动频率表征目标的机动情况。在应用当中，通常采用固定的机动频率，这就表示机动目标的机动时间是一定的，而实际上机动目标的机动时间是不断变化的，也就是说机动频率是不断变化的，采用固定机动频率必然会带来误差。采样周期在0.5—2S时，机动频率越小跟踪精度越高[1]，但机动频率仍然是固定值。本文提出的基于神经网络的机动频率自适应调整方法可以使机动频率随机动而变化，从而提高状态估计的准确性，提高跟踪精度。本文将小波神经网络用于机动目标跟踪中机动频率的自适应调整，该算法对机动目标“当前”统计模型中的机动频率进行实时修改, 从而自适应的改变机动频率,使跟踪算法与目标的真实状态更接近。该算法采用小波神经网络的离线训练，实时性好。-The maneuver of the maneuvering target is uncertain. The maneuvering frequency is constantly changeable, but traditionally it is beforehand determined as a constant based on the target state estimation in the state model of the maneuvering target. The maneuver of the maneuvering target makes the kinematics equation of the target model mismatch with the practical motion model and the tracking error will be increased. Based on the advantages of the self-learning, the rapid convergence rate and the nonlinear approximation ability of the wavelet neural network, it was put forward to be used in the field of target tracking in the paper. The new residual is used as the input of the wavelet neural network, the output of the network is used to adjust adaptively the maneuvering frequency of the CS model. The algorithm is more close to the real state of the target. The simulation results showed that tracking error can be reduced and the tracking accuracy can be improved. Platform: |
Size: 4096 |
Author:李隆基 |
Hits:
Description: 从改进提议分布的成片野值容错能力入手，提出了基于残差正交判别的UPF容错滤波算法，该算
法将残差正交判别法UKF的野值自适应性和粒子滤波的“适者生存性”有机地结合起来．通过非线性状态估计
的实验，证实了这种新的自适应粒子滤波对成片野值处理的有效性，-Proposal from the improved value of the distribution of fault tolerance into the film field, put forward an identification of the UPF-based fault-tolerant orthogonal residual filtering algorithm is orthogonal to the residual value of the wild Discriminant adaptive UKF and particle filter, " survival of the fittest of " organic integration. Nonlinear state estimation through experiments confirmed that this new adaptive particle filter into pieces on the effectiveness of treatment of outliers, Platform: |
Size: 163840 |
Author:陈洪 |
Hits:
Description: An implementation of Unscented Kalman Filter for nonlinear state estimation.-Nonlinear state estimation is a challenge problem. The well-known Kalman Filter is only suitable for linear systems. The Extended Kalman Filter (EKF) has become a standarded formulation for nonlinear state estimation. However, it may cause significant error for highly nonlinear systems because of the propagation of uncertainty through the nonlinear system.
The Unscented Kalman Filter (UKF) is a novel development in the field. The idea is to produce several sampling points (Sigma points) around the current state estimate based on its covariance. Then, propagating these points through the nonlinear map to get more accurate estimation of the mean and covariance of the mapping results. In this way, it avoids the need to calculate the Jacobian, hence incurs only the similar computation load as the EKF.
For tutorial purpose, this code implements a simplified version of UKF formulation, where we assume both the process and measurement noises are additive to avoid augment of state and a Platform: |
Size: 2048 |
Author:DT丿灬雪狼 |
Hits:
Description: 非线性滤波方法，主要包括EKF(扩展卡尔曼滤波)与UKF（无迹卡尔曼滤波），对于非线性状态、参数估计的学习有很大的帮助-Nonlinear filtering methods, including EKF (EKF) and UKF (unscented Kalman filter) for nonlinear state estimation is very helpful in learning Platform: |
Size: 729088 |
Author:夜思明 |
Hits:
Description: 本文讨论了小波神经网络在机动多目标跟踪中的应用,多目标跟踪就是主体为了维持对多个目标(客体)当前状态的估计而对所接收的量测信息进行处理的过程。以非线性大规模并行分布式处理为特征的神经网络可以解决传统的目标跟踪方法的难以解决的计算量组合爆炸问题以及需要确定机动目标的数学模型的问题, 将小波分析原理与神经网络相融合,提出了基于小波神经网络的目标跟踪方法来提高系统的学习能力、表达能力以及机动多目标状态的估计精度。-This article discusses the application of wavelet neural network in motorized multi-target tracking, multi-target tracking is the main measurement information received in order to maintain the current state of the multiple goals (guest) estimated processing. The nonlinear massively parallel distributed processing is characterized neural network can solve the traditional target tracking methods computational combinatorial explosion problem difficult to solve the mathematical model of the problem and the need to determine the maneuvering target, the principle of wavelet analysis and neural network fusion, target tracking method based on wavelet neural network to improve the system' s ability to learn, the ability to express and maneuvering multi-target state estimation accuracy. Platform: |
Size: 421888 |
Author:yaomeng |
Hits:
Description: 非线性扩展卡尔曼滤波算法的matlab程序-Description:This is a tutorial on nonlinear extended Kalman filter (EKF).
It uses the standard EKF fomulation to achieve nonlinear state estimation.
Inside, it uses the complex step Jacobian to linearize the nonlinear dynamic system.
The linearized matrices are then used in the Kalman filter calculation.
The complex step differentiation seems improving the EKF performance particularly in accuracy
such that the optimization and NN training through the EKF are better than through the UKF Platform: |
Size: 5120 |
Author:窦贤明 |
Hits:
Description: 基于非线性动力系统的无迹卡尔曼滤波matlab程序-onlinear state estimation is a challenge problem. The well-known Kalman Filter is only suitable for linear systems.
The Extended Kalman Filter (EKF) has become a standarded formulation for nonlinear state estimation.
However, it may cause significant error for highly nonlinear systems because of the propagation of uncertainty through the nonlinear system.
The Unscented Kalman Filter (UKF) is a novel development in the field Platform: |
Size: 8192 |
Author:窦贤明 |
Hits:
Description: H无穷控制针对马尔科夫调变系统的非脆弱控制研究。(This paper gives attention to the issue of
nonfragile state estimation for a class of Markov jump
systems with repeated scalar nonlinearities and redundant
channels.) Platform: |
Size: 357376 |
Author:如影骑士 |
Hits:
Search - nonlinear state estimation