Nonlinear Estimation Techniques Applied on Target Tracking Problems

2012 ◽  
Vol 134 (5) ◽  
Author(s):  
Andrew Gadsden ◽  
Saeid Habibi ◽  
Darcy Dunne ◽  
T. Kirubarajan
Keyword(s):  
Target Tracking ◽  
Traffic Control ◽  
Initial Conditions ◽  
Nonlinear Estimation ◽  
Variable Structure ◽  
Two Dimensions ◽  
Tracking Accuracy ◽  
Linear Measurements ◽  
Estimation Techniques ◽  
System Equations

This paper discusses the application of four nonlinear estimation techniques on two benchmark target tracking problems. The first problem is a generic air traffic control (ATC) scenario, which involves nonlinear system equations with linear measurements. The second study is a classical ground surveillance problem, where a moving airborne platform with a sensor is used to track a moving target. The tracking scenario is set in two dimensions, with the measurement providing nonlinear bearing-only observations. These two target tracking problems provide a good benchmark for comparing the following nonlinear estimation techniques: the common extended and unscented Kalman filters (EKF/UKF), the particle filter (PF), and the relatively new smooth variable structure filter (SVSF). The results of applying the SVSF on the two target tracking problems demonstrate its stability and robustness. Both of these attributes make use of the SVSF advantageous over other popular methods. The filters performances are quantified in terms of robustness, resilience to poor initial conditions and measurement outliers, and tracking accuracy and computational complexity. The purpose of this paper is to demonstrate the effectiveness of applying the SVSF on nonlinear target tracking problems, which in the past have typically been solved by Kalman or particle filters.

Target Tracking Using the Smooth Variable Structure Filter

2009 ◽  
Author(s):  
Andrew Gadsden ◽  
Saeid Habibi
Keyword(s):  
Target Tracking ◽  
Sliding Mode ◽  
Initial Conditions ◽  
A Priori ◽  
Variable Structure ◽  
Tracking Problem ◽  
Tracking Accuracy ◽  
State Estimate ◽  
Predictor Corrector ◽  
Smooth Variable Structure Filter

This article discusses the application of the smooth variable structure filter (SVSF) on a target tracking problem. The SVSF is a relatively new predictor-corrector method used for state and parameter estimation. It is a sliding mode estimator, where gain switching is used to ensure that the estimates converge to true state values. An internal model of the system, either linear or nonlinear, is used to predict an a priori state estimate. A corrective term is then applied to calculate the a posteriori state estimate, and the estimation process is repeated iteratively. The results of applying this filter on a target tracking problem demonstrate its stability and robustness. Both of these attributes make using the SVSF advantageous over the well-known Kalman and extended Kalman filters. The performances of these algorithms are quantified in terms of robustness, resilience to poor initial conditions and measurement outliers, tracking accuracy and computational complexity.

Locality-Sensitive Non-Linear Kalman Filter for Target Tracking

2021 ◽  
Vol 13 (1) ◽  
pp. 36-57
Author(s):  
Ben-Bright Benuwa ◽  
Benjamin Ghansah
Keyword(s):  
Kalman Filter ◽  
Target Tracking ◽  
Traffic Control ◽  
Unscented Kalman Filter ◽  
Research Area ◽  
Linear Filtering ◽  
Tracking Accuracy ◽  
Propagation Of Uncertainty ◽  
Non Linear ◽  
Marine Engineering

Target tracking (TT) with non-linear kalman filtering (NLKF) has recently become a very popular research area, particularly in the field of marine engineering and air traffic control. Contemporary NLKF algorithms have been very effective, in particular, with extensions and merging with a reduced root mean square error (RMSE) value. However, there are a number of issues that confront NLKF approaches, notably weakness in robustness, convergence speed, and tracking accuracy due to large initial error and weak observability. Furthermore, NLKF algorithms significantly results in error for high non-linear systems (NLS) because of the propagation of uncertainty. Again, there is a problem of estimating future states as a result of white noise. To handle these issues, the authors propose a novel non-linear filtering algorithm, called locality-sensitive NLKF (LSNLKF) that incorporates locality-sensitive adaptors into the structure of an integrated NLKF. They are the extended kalman filter (EKF) and the unscented kalman filter (UKF) for TT.

scholarly journals Target tracking accuracy and latency with different 4D ultrasound systems – a robotic phantom study

2020 ◽  
Vol 6 (1) ◽  
Author(s):  
Svenja Ipsen ◽  
Sven Böttger ◽  
Holger Schwegmann ◽  
Floris Ernst
Keyword(s):  
Target Tracking ◽  
Real Time ◽  
Template Matching ◽  
Image Data ◽  
Distinct Advantage ◽  
Reference Trajectory ◽  
Data Streaming ◽  
Tracking Accuracy ◽  
Time Data ◽  
4D Ultrasound

AbstractUltrasound (US) imaging, in contrast to other image guidance techniques, offers the distinct advantage of providing volumetric image data in real-time (4D) without using ionizing radiation. The goal of this study was to perform the first quantitative comparison of three different 4D US systems with fast matrix array probes and real-time data streaming regarding their target tracking accuracy and system latency. Sinusoidal motion of varying amplitudes and frequencies was used to simulate breathing motion with a robotic arm and a static US phantom. US volumes and robot positions were acquired online and stored for retrospective analysis. A template matching approach was used for target localization in the US data. Target motion measured in US was compared to the reference trajectory performed by the robot to determine localization accuracy and system latency. Using the robotic setup, all investigated 4D US systems could detect a moving target with sub-millimeter accuracy. However, especially high system latency increased tracking errors substantially and should be compensated with prediction algorithms for respiratory motion compensation.

An Improved Joint Target Tracking and Classification Algorithm Based on Data Fusion

2014 ◽  
Vol 904 ◽  
pp. 325-329
Author(s):  
Hong Wei Quan ◽  
Lin Chen ◽  
Dong Liang Peng
Keyword(s):  
Data Fusion ◽  
Target Tracking ◽  
Classification Algorithm ◽  
Target Classification ◽  
Tracking Accuracy ◽  
Traditional Methods ◽  
System Models ◽  
Target State ◽  
Classification Probability

This paper addresses the problem of the joint target tracking and classification based on data fusion. In traditional methods, a separate suite of sensors and system models are used, target tracking and target classification are usually treated as separate problems. In our JTC framework, the link between target state and class is considered and the feasibility of JTC techniques is discussed. The tracking accuracy and classification probability are improved to some extent with the more accurate classification results from classifier based on data fusion feedback to state filter.

Evolution of clusters of sedimenting low-Reynolds-number particles with Oseen interactions

2008 ◽  
Vol 603 ◽  
pp. 63-100 ◽  
Author(s):  
G. SUBRAMANIAN ◽  
DONALD L. KOCH
Keyword(s):  
Reynolds Number ◽  
Velocity Field ◽  
Aspect Ratio ◽  
Evolution Equation ◽  
Similarity Solution ◽  
Initial Conditions ◽  
Radial Distance ◽  
Two Dimensions ◽  
Number Density ◽  
Cluster Radius

A theoretical framework is developed to describe, in the limit of small but finite Re, the evolution of dilute clusters of sedimenting particles. Here, Re =aU/ν is the particle Reynolds number, where a is the radius of the spherical particle, U its settling velocity, and ν the kinematic viscosity of the suspending fluid. The theory assumes the disturbance velocity field at sufficiently large distances from a sedimenting particle, even at small Re, to possess the familiar source--sink character; that is, the momentum defect brought in via a narrow wake behind the particle is convected radially outwards in the remaining directions. It is then argued that for spherical clusters with sufficiently many particles, specifically with N much greater than O(R0U/ν), the initial evolution is strongly influenced by wake-mediated interactions; here, N is the total number of particles, and R0 is the initial cluster radius. As a result, the cluster first evolves into a nearly planar configuration with an asymptotically small aspect ratio of O(R0U/N ν), the plane of the cluster being perpendicular to the direction of gravity; subsequent expansion occurs with an unchanged aspect ratio. For relatively sparse clusters with N smaller than O(R0U/ν), the probability of wake interactions remains negligible, and the cluster expands while retaining its spherical shape. The long-time expansion in the former case, and that for all times in the latter case, is driven by disturbance velocity fields produced by the particles outside their wakes. The resulting interactions between particles are therefore mutually repulsive with forces that obey an inverse-square law. The analysis presented describes cluster evolution in this regime. A continuum representation is adopted with the clusters being characterized by a number density field (n(r, t)), and a corresponding induced velocity field (u (r, t)) arising on account of interactions. For both planar axisymmetric clusters and spherical clusters with radial symmetry, the evolution equation admits a similarity solution; either cluster expands self-similarly for long times. The number density profiles at different times are functions of a similarity variable η = (r/t1/3), r being the radial distance away from the cluster centre, and t the time. The radius of the expanding cluster is found to be of the form Rcl (t) = A (ν a)1/3N1/3t1/3, where the constant of proportionality, A, is determined from an analytical solution of the evolution equation; one finds A = 1.743 and 1.651 for planar and spherical clusters, respectively. The number density profile in a planar axisymmetric cluster is also obtained numerically as a solution of the initial value problem for a canonical (Gaussian) initial condition. The numerical results compare well with theoretical predictions, and demonstrate the asymptotic stability of the similarity solution in two dimensions for long times, at least for axisymmetric initial conditions.

scholarly journals Research on Target Tracking Algorithm Based on Siamese Neural Network

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Haibo Pang ◽  
Qi Xuan ◽  
Meiqin Xie ◽  
Chengming Liu ◽  
Zhanbo Li
Keyword(s):  
Neural Network ◽  
Target Tracking ◽  
Evaluation Criteria ◽  
Attention Mechanism ◽  
Tracking Algorithm ◽  
Temporal Attention ◽  
Tracking Accuracy ◽  
Current Frame ◽  
Tracking Process ◽  
Recognition Ability

Target tracking is a significant topic in the field of computer vision. In this paper, the target tracking algorithm based on deep Siamese network is studied. Aiming at the situation that the tracking process is not robust, such as drift or miss the target, the tracking accuracy and robustness of the algorithm are improved by improving the feature extraction part and online update part. This paper adds SE-block and temporal attention mechanism (TAM) to the framework of Siamese neural network. SE-block can refine and extract features; different channels are given different weights according to their importance which can improve the discrimination of the network and the recognition ability of the tracker. Temporal attention mechanism can update the target state by adjusting the weights of samples at current frame and historical frame to solve the model drift caused by the existence of similar background. We use cross-entropy loss to distinguish the targets in different sequences so that their distance in the feature domains is longer and the features are easier to identify. We train and test the network on three benchmarks and compare with several state-of-the-art tracking methods. The experimental results demonstrate that the algorithm proposed is superior to other methods in tracking effect diagram and evaluation criteria. The proposed algorithm can solve the occlusion problem effectively while ensuring the real-time performance in the process of tracking.

Research on Underwater Target Tracking Based on Contour Detection

2011 ◽  
Vol 317-319 ◽  
pp. 890-896
Author(s):  
Ming Jun Zhang ◽  
Yuan Yuan Wan ◽  
Zhen Zhong Chu
Keyword(s):  
Prediction Model ◽  
Target Tracking ◽  
Tracking System ◽  
Contour Detection ◽  
Motion Prediction ◽  
Moving Target ◽  
Contour Extraction ◽  
Tracking Accuracy ◽  
Tracking Method ◽  
Underwater Target

The traditional centroid tracking method over-relies on the accuracy of segment, which easily lead to loss of underwater moving target. This paper presents an object tracking method based on circular contour extraction, combining region feature and contour feature. Through the correction to circle features, the problem of multiple solutions causing by Hough transform circle detection is avoided. A new motion prediction model is constructed to make up the deficiency that three-order motion prediction model has disadvantage of high dimension and large calculation. The predicted position of object centroid is updated and corrected by circle contour, forming prediction-measurement-updating closed-loop target tracking system. To reduce system processing time, on the premise of the tracking accuracy, a dynamic detection method based on target state prediction model is proposed. The results of contour extraction and underwater moving target experiments demonstrate the effectiveness of the proposed method.

Sliding Mode Control of a Three Degrees of Freedom Anthropoid Robot by Driving the Controller Parameters to an Equivalent Regime

2000 ◽  
Vol 122 (4) ◽  
pp. 632-640 ◽  
Author(s):  
M. Onder Efe ◽  
Okyay Kaynak ◽  
Xinghuo Yu
Keyword(s):  
Degrees Of Freedom ◽  
Sliding Mode ◽  
Initial Conditions ◽  
Tracking Error ◽  
Variable Structure ◽  
Mathematical Representation ◽  
Dynamic Complexity ◽  
Variable Structure Systems ◽  
Sliding Motion ◽  
Three Degrees Of Freedom

Noise rejection, handling the difficulties coming from the mathematical representation of the system under investigation and alleviation of structural or unstructural uncertainties constitute prime challenges that are frequently encountered in the practice of systems and control engineering. Designing a controller has primarily the aim of achieving the tracking precision as well as a degree of robustness against the difficulties stated. From this point of view, variable structure systems theory offer well formulated solutions to such ill-posed problems containing uncertainty and imprecision. In this paper, a simple controller structure is discussed. The architecture is known as Adaptive Linear Element (ADALINE) in the framework of neural computing. The parameters of the controller evolve dynamically in time such that a sliding motion is obtained. The inner sliding motion concerns the establishment of a sliding mode in controller parameters, which aims to minimize the error on the controller outputs. The outer sliding motion is designed for the plant. The algorithm discussed drives the error on the output of the controller toward zero learning error level, and the state tracking error vector of the plant is driven toward the origin of the phase space simultaneously. The paper gives the analysis of the equivalence between the two sliding motions and demonstrates the performance of the algorithm on a three degrees of freedom, anthropoid robotic manipulator. In order to clarify the performance of the scheme, together with the dynamic complexity of the plant, the adverse effects of observation noise and nonzero initial conditions are studied. [S0022-0434(00)01704-4]

Sign in / Sign up

Export Citation Format

Share Document