Word Files
Reference for Downloading Educational Files
  2. Electrical Engineering
  3. Improvement of GM-PHD algorithm for multi-target and multi-sensor tracking with the help of bias estimation

Improvement of GM-PHD algorithm for multi-target and multi-sensor tracking with the help of bias estimation

Number of pages: 105 File Format: word File Code: 32180
Year: Not Specified University Degree: Master's degree Category: Electrical Engineering
Tags/Keywords: Algorithm improvement - bias - Bias estimation - GM-PHD algorithm - GM-PHD filter - Multi-target tracking - PhD - Probability density hypothesis - Sensor
  • Part of the Content
  • Contents & Resources

  • Summary of Improvement of GM-PHD algorithm for multi-target and multi-sensor tracking with the help of bias estimation

    Dissertation for receiving the master's degree "M.Sc. »

    Field of study: Electricity

    Inclination: Telecommunication

    Abstract

    In this thesis, multi-objective and multi-sensor tracking problems are investigated using the random particle set rule. The regression probability density hypothesis (PHD) is implemented dynamically. This dynamic mode is performed by combining the transfer bias value with the intensity function. We have assumed the dynamic bias as linear Gaussian in the intensity function. Gaussian filter implementation is closed form Gaussian components. The target position and the transfer bias value are coupled by the accuracy function in each step. Using the two-stage Kalman filter leads to a significant reduction in the complexity of calculations. Here are two examples to check the proposed filter.

    Keywords:

    multi-target tracking, probability density hypothesis filter, bias estimation

    Introduction

    Since tracking targets using Data received from a sensor has constant error and noise. Several sensors can be used to reduce these errors. But using multiple sensors has the problem of bias error not being the same. that this error is combined with the actual value of the target position and as a result it is not possible to estimate the correct position of the targets by means of several sensors in common coordinates. Therefore, to reduce the sensor error, we need to estimate the bias value of each sensor, and finally, by subtracting the bias error from the data value measured by the sensor, we estimate the correct position of the target in the joint coordinates. Here we examine the upcoming problems in target tracking using multiple sensors and bias estimation. 1-1 Statement of the problem In this thesis, multi-target and multi-sensor tracking problems are investigated using the random particle set rule. The recursive probability density theorem (PHD[1]) is implemented dynamically, and this dynamic mode is performed by combining the transfer bias value with the intensity function. GM-PHD filter has the ability to estimate the number and status of targets, based on noisy observations and in the presence of false targets. However, in collision situations where the targets cross over each other, the GM-PHD filter faces problems and loses its efficiency. On the other hand, compensating the error recorded in the integrity of the data received from several sensors is an important issue, regardless of whether their measurement is centralized or distributed.

    There are different methods for biasing sensors, for example:

    transitional bias

    bias Rotational

    Transitional and rotational bias (Sudano, 1993).

    Here we seek to choose the appropriate bias in terms of response speed and accuracy in tracking multiple targets, by receiving data from several sensors in 2D coordinates with the help of a PHD filter.

    The problem of estimating the size of the unknown bias is of interest Many have been placed. If it is possible to estimate the bias value correctly, multi-sensor measurements can be used in the form of joint coordinates. In Figure 1?1, you can see the measurement of several sensors without recording errors

    The triangles indicate the location of each sensor in common coordinates, the circles indicate the original position of the targets, and the squares indicate the measurements produced by each sensor.

    To solve the bias estimation problem unknown, various methods have been proposed.

    Probability method[1]

    Kalman filter method

    Of these methods, the Kalman filter method has received a lot of attention.

    Although it is possible to implement by ASKF[3] from the point of view of practical and numerical calculations, it causes problems such as ill-posedness[4]. To reduce this complexity Friedland [5], two-stage estimation by separating the bias from the target position can be proposed (Friedland, 1969). When there is a special relationship between the initial parameters of the 2 filters, this two-stage estimation is equal to the one-stage ASKF estimation (Ignagni, 1981).

    It should be noted here that the problem of many existing methods is the uncertainty in the source measurement, which often occurs in multi-target tracking. Although techniques such as joint data probability [6] and multi-probability tracking [7] that have been obtained so far can be used. But they may not have a favorable result due to not considering the bias effect. Recently, the set of statistical particle set theory [8] has been used to deal with multi-objective tracking problems in data communication (Mahler, MA, 2007). The structure of the set of statistical particles is such that it models the position of the target and the bias value as two finite random sets [9], and as a result, the problem of tracking unknown targets at different times in environments with interference is naturally solved. In addition, multi-objective tracking can be expressed in the Bayesian framework by creating a transfer density and a multi-objective accuracy function [10]. Probability hypothesis density (PHD) dynamic running recessive. The dynamic state of the modulation transfer function of the bias intensity reduction. Under the linear Gaussian assumptions on the bias dynamics, the Gaussian mixture implementation is used to give closed-form expressions. As the target state and the translational measurement bias are coupled through the likelihood in the update step, the use of two-stage Kalman filter leads to a significant reduction in the computational complexity. Two examples are provided to verify the proposed filter.

  • Contents & References of Improvement of GM-PHD algorithm for multi-target and multi-sensor tracking with the help of bias estimation

    List:

    Title

    Abstract. 1

    Introduction.. . 2

    Chapter One: Generalities. 3

    1-1 Statement of the problem. 4

    1-2    research objectives. 8

    1-3    Hypotheses 8

    1-4    Research background. 8

    1-5    Research method. 9

    Chapter Two: Research background. 10

    Introduction.. 11

    2-1 Multi-objective tracking model by Bayesian filter. 11

    2-2 Gaussian filter. 13

    2-2-1 Multi-objective tracking model by PHD filter. 14

    2-3 Monte Carlo filter. 22

    2-3-1 Ordinal Monte Carlo. 23

    2-4    SMC-PHD filter with error registration 30

    2-4-1     Investigation of the error registration problem 34

    2-4-2     SMC-PHD simulation with error registration 36

    Chapter three: GM-PHD with the help of bias estimation. 44

    Introduction    45

    3-1    GM-PHD filter with the help of bias estimation for linear targets. 50

    3-1-1 First step: prediction. 50

    3-1-2     Second step: update. 51

    3-1-3     The third stage: Pruning and integration of Gaussi members. 56

    3-1-4 Fourth step: target position estimation and sensor bias estimation 60

    3-2 GM-PHD filter with the help of bias estimation for tracking non-linear (maneuvering) targets 61

    3-2-1 First step: BFG approximation. 61

    3-2-2 The second stage: prediction. 65

    3-3 Evaluation criteria of filter types. 66

    3-4    PHD error convergence. 68

    3-5    Implementation of GM-PHD filter with the help of bias estimation. 73

    3-5-1     GM-PHD implementation algorithm with the help of bias estimation for linear targets. 73

    3-5-2     GM-PHD implementation algorithm with the help of bias estimation for non-linear targets. 74

    Chapter Four: Simulation. 75

    Introduction 76

    4-1 Simulation 1. 76

    4-2 Simulation 2. 85

    Chapter Five: Conclusion and Suggestions 94

    5-1 Conclusion. 95

    2-5 Suggestions 98

    Resources. 99

    English abstract. 1

    Source:

     

    - Alspach, D. (1970). A Bayesian approximation technique for estimation and control of discrete time systems. Univ. Calif.

    - Anderson, B., & Moore, J. (1979). Optimal Filtering. Englewood Cliffs.

    - Beddey, A., & van Lieshout, M. (1992). ICM for object recognition. Springer.

    - Clark, D., & Bell, J. (Jul.2006). Convergence results for the particle PHD filter. IEEE Trans. Signal Process, 54(7), 2652–2661.

    - Clark, D., & Bell, J. (Oct. 2005). Bayesian multiple target tracking in forward scan sonar images using the PHD filter. Proc. Inst. Elect. Eng.—Radar, Sonar, Navigation, 152(5), 327–334.

    - Clark, D., & Vo, B. (2007). Convergence analysis of the Gaussian mixture PHD filter. IEEE Transactions on Signal Processing, 55(4), 1204–1211.

    - Clark, D., Panta, K., & Vo, B. (Jul 2006). The GM-PHD filter multiple target tracker. Information Fusion. Florence, Italy.

    - Clark, D., Ruiz, I., Petillot, Y., & Bell, J. (2007). Particle PHD filter multiple target tracking in sonar image. IEEE Trans. Aerosp. Electron. Syst, 43(1), 409–416.

    - Clark, D., Vo, B., & Bell, J. (Apr 2006). GM-PHD filter multi-target tracking in sonar images. presented at the SPIE Defense Security Symp, 17–21.

    - Dana, M. (MA 1990). Registration: a prerequisite for multiple sensor tracking. Multitarget multisensor tracking. In advanced applications (pp. 155–185). Norwood: ArtechHouse Publishers,.

    - Doucet, A., De Freitas, N., & Gordon, N. (May 2001). Sequential Monte Carlo Methods in Practice. Springer Springer-Verlag.

    - El-Fallah, A., & Mahler, R. (May 2011). Bayesian unified registration and tracking. Proceedings of the SPIE Conference on Signal Processing, Sensor Fusion and Target Recognition, 8050, 1–11.

    - Friedland, B. (1969). Treatment of bias in recursive filtering. IEEE Transactions on Automatic Control 14 (4), 14(4), 359–367.

    - Geyer, C. (1999). Likelihood inference for spatial pointLikelihood inference for spatial point processes. Stochastic Geometry, probability and computation, 79-140.

    - Goodman, I., Mahler, R., & Nguyen, H. (1997). Mathematics of Data Fusion. Kluwer Academic Publishers.

    - Grimmett, G., & Stirzaker, D. (2011). One Thousand Exercises in Probability. London: Oxford University Press.

    - Hernandez, M., Ristic, B., Farina, A., & Sathyan, T. (2008). Performance measure for Markovian switching systems using best fitting Gaussian distributions. IEEE Transactions on Aerospace and Electronic Systems, 44 (2), 724–747.

    - Herrero, J., Portas, J., & Corredera, J. (2007). On-line multi-sensor registration for data fusion on airport surface. IEEE Trans. Aerosp. Electron. Syst, 43(1), 356–370.

    - Ignagni, M. (1981). An alternate derivation and extension of Friedland's two-stage Kalman estimator. IEEE Transactions on Automatic Control, 26(3), 746–750.

    - Ikoma, N., Uchino, T., & Maeda, T. (Aug 2004). Tracking of feature points in image sequence by SMC implementation of PHD filter. in Soc. Instrument and Control Engineers (SICE) 2004 Annu. Conf, 2, 1696–1701.

    - Johansen, A., Singh, S., Doucet, A., & -N, B. (Jun. 2006). Convergence of the SMC implementation of the PHD filter. Method. Comput. Appl. Probab, 8(2), 265–291.

    - Julier, S., & Uhlmann, J. (1996). A general method for approximating nonlinear transformations of probability distributions. RRG, Eng. Sci. Dept., Univ. Oxford, Oxford, U.K.,.

    - Julier, S., & Uhlmann, J. (1997). A new extension of the Kalman filter to nonlinear systems. in Int. Symp. Aerosp./Defense Sensing, Simultaneous Controls, Orlando, FL.

    - Li, X., & Jilkov, V. (April 2000). A survey of maneuvering target tracking: dynamic models. Proc. 2000 SPIE Conf. on Signal and Data Processing of Small Targets, 4048, 212–235.

    - Lian, F., Han, C., Liu, W., & Chen, H. (2011). Joint spatial registration of a multi-target tracking using an extended probability hypothesis density filter. IET Radar, Sonar and Navigation, 5(4), 441–448.

    - Lo, H. (1972). Finite-dimensional sensor orbits and optimal non-linear filtering. IEEE Trans. Inf. Theory, IT-18(5), 583–588.

    - Ma, W., Singh, S., & Vo, B. (2004). Tracking multiple speakers with random sets. Proceedings of the International Conference on Acoustics, Speech and Signal Processing, (pp. 357–360). Montreal, Canada.

    - Mabler, R. (June 2000). A theoretical foundation for the Stein-Wmter Probability Hypothesis Density (PHD) multi-target tracking approach. Pmc.2002 MSS Nat? Symp. on Sensor and Data Fusion, 1. San Antonio TX.

    - Maggio, E., Taj, M., & Cavallaro, A. (2008). Efficient multitarget visual tracking using random finite sets. IEEE Trans. Circuits Syst. Video Technol, 18(8), 1016–1027.

    - Mahler, R. (1994). Global integrated data fusion. Proc. 7th Nat. Symp. on Sensor Fusion, 1, 187-199.

    - Mahler, R. (2000). Approximate multisensor-multitarget joint detection, tracking and identification using a first order multitarget moment statistic. IEEE lhm. AES.

    - Mahler, R. (2003). Multi-target Bayes filtering via first-order multi-target moments. IEEE Trans. Aerosp. Electron. Syst, 39(4), 1152–1178.

    - Mahler, R. (2010). Approximate multisensor CPHD and PHD filters. Proceedings of the 13th International Conference on Information Fusion, 1-8.

    - Mahler, R. (MA, 2007). Statistical Multisource-Multitarget Information Fusion. Artech House, Norwood,.

    - Mahler, R. (March 2000). An Introduction to Multisource-Multitarget Statistics and Applications. Lockheed Martin Technical Monograph.

    - Nagappa, S., & Clark, D. (May 2011). On the ordering of the sensors in the iterated-probability hypothesis density (PHD) filter. Proceedings of the SPIE Conference on Signal Processing, Sensor Fusion and Target Recognition, 8050, 1-6.

    - Okello, N., & Ristic, B. (2003).

Improvement of GM-PHD algorithm for multi-target and multi-sensor tracking with the help of bias estimation
How To Access The File
Related Contents:
Number of pages: 120 Category: Electronic Engineering

Field of study: Electricity, specialization: Telecommunication. Dissertation Abstract In this thesis, the problems of multi-objective and multi-sensor tracking are investigated using the rule of random particle collection. The regression probability density hypothesis (PHD) is implemented dynamically. This dynamic mode is performed by combining the transfer bias value with the ...

Number of pages: 129 Category: Geography - Urban Planning

Dissertation for M.Sc Master's Degree in Remote Sensing and Geographical Information System - Water and Soil Resources Abstract One of the important issues in crisis management of unexpected natural events, especially earthquakes, is optimal location in order to accommodate citizens during or after an accident. One of the major problems of the organizations involved in urban ...

Number of pages: 99 Category: Facilities - Mechanics

Dissertation for the degree of Master of Mechatronics M.SC. Abstract: The subject of this research is the design and simulation of data integration in a radar network that has overlapping. Data fusion means combining the output data of dissimilar radar sensors that are different in terms of accuracy in ranging and angle measurement. These sensors are installed in a row in front ...

Number of pages: 92 Category: Electronic Engineering

Dissertation for master's degree, specialization: Power, Power, Abstract One of the important design challenges in wireless sensor networks (WSN) is to prolong the lifetime of the system (node), while achieving an acceptable quality of service for the applications. Extending the life of the node is directly related to reducing energy consumption. In WSN, each sensor node works ...

Number of pages: 104 Category: Computer Engineering

Master's Thesis of Computer Engineering - Computer Architecture Abstract With the advancement of electronic device manufacturing technology and the cost-effectiveness of large-scale sensor networks, wireless sensor networks provide research fields with rapid growth and great interest, which have attracted a lot of attention in recent years. Large-scale wireless sensor networks ...

Number of pages: 67 Category: Computer Engineering

Master's Thesis in Computer Engineering (M.Sc) Abstract: Today, automatic control is widely used in transportation systems. There are several reasons for using this method, the most important of which are cost reduction, the ability to make devices lighter, more safety for the passengers or to keep them away from danger. The last two cases are more visible in systems that have ...

Number of pages: 120 Category: Electronic Engineering

Master's thesis in the field of electrical engineering-communication system for removing interference in the passive radar reference channel based on digital TV broadcast signal by terrestrial transmitters with a reproduction approach. In this thesis, a digital receiver for signal processing in a passive radar reference receiver based on orthogonal frequency division modulation ...

Number of pages: 112 Category: Electrical Engineering

Master's Thesis in the field of Electrical Engineering-Communication System, abstract of the investigation of the methods of removing interference signals in the surveillance channel in passive radars based on DVB-T signal. In this thesis, the performance of adaptive methods in weakening the interference signals including the direct path signal and clutter and the degree of ...

Number of pages: 96 Category: Electronic Engineering

Master's Thesis in Electronics Abstract Nowadays, according to the advantages of wireless sensor networks, which are simple and cheap implementation, low power consumption and high scalability, they are used in many applications. Designing sustainable wireless sensor networks is a very challenging issue. Sensors with limited energy are expected to operate automatically for a ...

Number of pages: 80 Category: Electronic Engineering

Master's Thesis in Electronics Abstract Nowadays, according to the advantages of wireless sensor networks, which are simple and cheap implementation, low power consumption and high scalability, they are used in many applications. Designing sustainable wireless sensor networks is a very challenging issue. Sensors with limited energy are expected to operate automatically for a ...