Instituto de Sistemas e Robótica * Pólo de Coimbra * Tracking and Data Association Using Data from a LRF

Size: px
Start display at page:

Download "Instituto de Sistemas e Robótica * Pólo de Coimbra * Tracking and Data Association Using Data from a LRF"

Transcription

1 Instituto de Sistemas e Robótica * Pólo de Coimbra * Tracing and Data Association Using Data from a LRF Technical Report Nº ISRLM2005/04 Cristiano Premebida October, Departamento de Engenharia Electrotécnica e de Computadores, Pólo II Universidade de Coimbra Coimbra - PORTUGAL

2 Instituto de Sistemas de Robótica Pólo de Coimbra Ano: 2005 * Laboratório de Mecatrónica * TR. Nº.: ISRLM2005/04 Tracing and Data Association Using Data from a LRF * Technical Report * Editors: Cristiano Premebida 1 Prof. Urbano Nunes Tas position: Msc. Student Scientific Advisor This wor is supported by Institute for Systems and Robotics ISR, the Portuguese Science and Technology Foundation (FCT) under grant: POSC/EEA-SRI/58279/2004, and project CyberC3: CN/ASIA-IT&C/002 (88667). 1 CPremebida is supported by the Programme Alβan, the European Union Programme of High Level Scholarships for Latin America, scholarship nº E04M029876BR Institute of Systems and Robotics ISR, Coimbra, Portugal. Local: Coimbra, Portugal Date: Authors: CPremebida

3 TRACKING AND DATA ASSOCIATION USING DATA FROM A LRF Cristiano Premebida UC/FCT/DEEC-MEEC Abstract This report presents some results and experimental evaluations on Tracing and Data Association approaches using data from a SICK/LMS200 laser sensor. Some filters are briefly described (KF, EKF, PDAF and PF) and results presented. The target (object of interest) considered in this wor is assumed to move in a 2-D space (an outdoor environment) and its motion is modeled as independent between coordinates. The motion in each coordinate has nearby constant velocity with piecewise constant white noise acceleration. Keywords Laser scanning; Kalman filters; Detection; Tracing; Data Association. Instituto de Sistemas de Robótica - ISR 3 of 17

4 Appendix 1 Introduction and Bacground Objectives Tracing of a single non-stationary target KF technique EKF Technique Standard Particle Filter situation PDAF method References Instituto de Sistemas de Robótica - ISR 4 of 17

5 1 Introduction and Bacground Some useful concepts, that will be the basis of this report, are stated bellow: Estimation: the process of inferring the value of a variable of interest from indirect, inaccurate and uncertain observations (process of measurement). Tracing: is the estimation of the state of a dynamic process (Object of interest). This is done using the tools from estimation theory and also statistical decision theory, especially when some problems (e.g. data association) are considered. Data Association: process of establishing which measurement (or weighted combination of measurement in some Validation Region) is to be used in a state estimator. Filtering: is the estimation of the current state of a dynamic process from noisy data. This report presents some results and experimental evaluations on Tracing and Data Association approaches using data from a SICK/LMS200 laser sensor. The framewor of this report, which bloc diagram is illustrated on Fig-1, was done in the context of the MTDS Project (Multi-Target Detection and Tracing System): Figure 1: Bloc Diagram of the System 2 Objectives The objectives of this text are to present some results, performed mainly in MatLab computational environment, for the tracing and data association problem relying on: - Basic tracing algorithms (KF and EKF); - Basic Particle Filter techniques; - Probabilistic Data Association Filter (PDAF). The situations of interest in this wor are: - Tracing and data association of single targets; - Tracing targets with stationary (non-maneuvering) behavior; - Tracing targets with maneuvering behavior; - Tracing of nearby targets (Cluster situation); Instituto de Sistemas de Robótica - ISR 5 of 17

6 Future wor will be focused on: - Multi-targets situation - Improvement on the techniques listed above - Cooperative sensor fusion: using the Laser and a single camera - Data association techniques in complex (realistic) situations 3 Tracing of a single non-stationary target In order to maintain a properly estimate of the process state (the tracing problem) it is crucial to carry out the data correlation process (data association) problem. Data association problem may be categorized according to what is associated what; in our case we have interest in measurement to trac association category, i.e. trac maintenance or updating. As described in [1], there are two fundamentally different approaches in associating the data: - Non-Bayesian data association: based on a decision procedure, using some statistical tools, a data association decision is made without consideration on the fact that this association is not necessarily correct. - Probabilistic (Bayesian) data association: this approach evaluates probabilities of associations and uses them throughout the estimation process. Situations of interest: I. Trac maintenance with single sensor without measurement uncertainty Non-maneuvering targets (stationary) - KF: with only uncertainty being white (Gaussian) noise with nown moments - KF: models with assumed zero-mean acceleration (jer) white noise - EKF: system/measurement are nonlinear Maneuvering targets (dynamic) - KF/EKF: with variable process noise - VSD (variable state dimension): switching from low order quiescent model to higher order maneuvering model - IE (input estimation): the unnown input (maneuver) is estimated under the assumption that it is constant over a window and the state estimate is corrected with it - MM (multiple model): II. Trac maintenance with single sensor with measurement uncertainty Single target (non-maneuvering) in clutter - Nearest (or strongest) neighbor standard KF (NNSKF): the measurement nearest to the predicted location or the strongest detection in the gate (validation region) are used as the correct one; - PDAF (Probabilistic Data Association Filter): use of all the measurements in the current time validation region, weighted by the computed association probability. Tracing with multiple sensors and/or multiples targets will not be treated in this report. Instituto de Sistemas de Robótica - ISR 6 of 17

7 The experimental results are based on the tracing of a moving object which dynamic behavior, considered here as a maneuvering model, is governed by a mobile robot (Robucar) performing a quasi circular trajectory on an outdoor environment. Incoming measurements data come from a SICK-LMS200 active laser sensor mounted in a stationary position. The variable of interest in our estimation case is the state of a dynamic system, which in our context consist of: - cinematic components: position and velocity - feature components: geometrical parameters, target classification The application of standard estimation algorithms, which would be used the measurement nearest in some sense to the predicted measurement nearest neighbor approach can lead to very poor results in an environment where spurious measurement occurs frequently. This is because such an approach does no account for the fact that measurement used in the algorithm might be originated from a source different from the target of interest. In addiction to their inaccuracy, the measurement system can be affected by additional uncertainty which is usually modeled by some additive noise. This problem (this additional uncertainty) is related to the origin of the measurement; this situation can occur in systems operating in the presence of - clutter (due to spurious reflectors or radiators near the target of interest) - decoys and countermeasures - random false alarms in the detection process - interfering targets The measurements of interest are not the raw data points from the Laser, but the outputs of the Segmentation and Feature Extraction blocs. In the sequel, we will present some results for the tracing of a single object (target). The modeling process will be carried out using two approaches: - Standard linear model - Non-linear standard model - Expanded model: acceleration component is added as a state to be estimated System model Considering the discrete-time controlled process that is governed by the linear stochastic difference equation x ω, = Ax 1 + Bu + 1 x n R (1) with measurements modeled by z = + ν, Hx z m R (2) Instituto de Sistemas de Robótica - ISR 7 of 17

8 The random variables ω and ω represent the process and measurement noise (respectively), and are assumed to be zero-mean, white and independent, with normal probability distributions: p( ω) N(0, Q) p( ν ) N(0, R) (3) The symbol N(x,µ,S) stands for the normal (Gaussian) pdf with arguments: the (vector) random variable x, mean µ, and covariance S. In this wor, the motion of a target (object of interest) is best described in Cartesian coordinates, with the state-space vector: x p p = v v x / y / x / y / (4) where: - p x : geometrical center position of the object related to X-axis - p y : geometrical center position of the object related to Y-axis - v x : linear velocity in direction of the X axis - v y : linear velocity in direction of the Y axis Target model The target is assumed to move in a 2-Dimensional space and its motion is modeled as independent between coordinates. The motion in each coordinate has nearby constant velocity with piecewise constant white noise acceleration. The measurements are position only (range,bearing), with additive white noise with variance (σ r ) in each coordinate. Variables on consideration: Polar coordinates True range/bearing: Measured range/bearing: Error*: *_ assumed independent, with zero-mean and standard deviation σ r, σ θ. Cartesian coordinates True range/bearing: Measured range/bearing: Error: - exact error: - linearized error: Without necessity of demonstration, the straightforward equations define the variables stated before: Instituto de Sistemas de Robótica - ISR 8 of 17

9 The following covariance matrices will be used on simulations: r + mσθ sin θm σ r cos θm ( σ r rmσ θ )sinθm cosθm RKF = ( σ r rmσ θ )sinθm cosθm rmσ θ cos θm + σ r sin θm (5) 2 σ 0 R = r EKF 2 0 σ θ (6) 3.1 KF technique The well nown fundamental assumptions of the KF are: 1- Evolution of the system s state according to a nown linear process equation driven by: - A nown input - An additive process noise, zero-mean white (uncorrelated) process with nown covariance Q(). 2- Measurements - a nown linear function of the state with - An additive measurement noise, zero-mean white with nown covariance R() 3- Initial state unnown, assumed to be a random variable with a nown mean (initial) and covariance (initial uncertainty) 4- Initial error and noises, mutually uncorrelated As almost all recursive algorithms, the state estimation cycle consists of I- Prediction: of the state and measurements (time update) II- Update: of the state (measurement update) Remars Under the Gaussian assumption for the initial state error and all the noises entering into the system, the KF is the MMSE state estimator. If the above random variables are not Gaussian, and one has only their first two moments, then the KF is the best linear MMSE estimator (in the context/class of linear estimator). The covariance calculations are independent of the state, and can be performed offline. Limitations on using KF - nonlinear motion models - nonlinear measurement equations - unnown inputs into the dynamic equation and/or mode changes - auto or cross-correlated noises - uncertain origin measurements (data association uncertainty) - unnown number of multiple-targets Considering the inputs equal zero, i.e. u =0, the state prediction equations on X- direction are given by: p v x / x / 1 = 0 t p 1 v x / 1 x / 1 ω px + ωvx / 1 / 1 (7) with the measurement Instituto de Sistemas de Robótica - ISR 9 of 17

10 px / = [ 1 ] + vx (8) z px / 0 / v x / In a similar manner, the Y-direction equations: p v y / y / 1 = 0 t p 1 v y / 1 y / 1 ω py + ωvy / 1 / 1 (9) where: py / z py / = [ 1 0] + vy / v (10) y / Results The object of interest to be traced is assumed having a constant linear velocity motion, executing a semi-circular trajectory in an outdoor environment. The measurements are done in polar coordinates (r,θ) using the Laser sensor and we have interest in tracing the Cartesian form, thus r cos( θ ) z polar = ( r, θ ) zcartesinan = (11) r sin( θ ) This transformation avoids the use of a nonlinear function model. Fig-2 presents the practical experiment on estimating/updating the Cartesian position of the Object over time KF - Tracing the position of a moving object Estimated Measurement Position - Y Position - X Figure 2: Tracing the position of the target using a KF Instituto de Sistemas de Robótica - ISR 10 of 17

11 3.2 EKF Technique Modeling assumptions (the same as for the KF) - The dynamic and/or the measurement equation are nonlinear functions Based on first or second order Taylor expansion series of the nonlinear functions (in the prediction stage), we have the 1 st or 2 nd order EKF. The use of the series expansion in the state/measurement prediction has the undesired potential of introducing unmodeled errors that violate some basic assumptions about the prediction errors. In general, a nonlinear transformation will - introduce a bias (in contrast with the zero-mean consideration), and - the covariance calculation based on a series expansion is not always accurate. The measurements are done in polar coordinates, i.e. the nonlinearities are postpone to the Filter, thus z ] T = [ r, θ (12) These expansions rely on Jacobian/Hessians evaluated at the estimated state (rather then the exact state) and neglect the higher order terms. By the way, the EKF are very sensitive to the accuracy of the initial conditions. Nevertheless there are some techniques to compensate these undesired situations, for example: using an IEKF, using artificial noise, etc [1]. In this case, we have the same equations, as the KF case, for the state prediction but, the measurement prediction is the nonlinear mapping: z = h( x, ν, ) (13) where 2 2 1/ 2 ( p + x py ) h = (14) arctan( py / px) As stated before, our state-space vector is represented by x p p = v v x / y / x / y / Results The nonlinearities are treated in a suboptimal way using the Jacobian (1 st order EKF) of the function h evaluated at the predicted state estimate, thus Instituto de Sistemas de Robótica - ISR 11 of 17

12 h1( + 1) h1 ( + 1) K x1 x2 h2 ( + 1) h2 ( + 1) H = h = K x1 x (15) 2 M O M x = xˆ ( ) In our case, the measurement covariance update is 2 2 1/ / 2 x1 ( x1 + x3 ) 0 x3( x1 + x3 ) 0 H = (16) 2 x3x1 [(1 + ( x3 / x1 ) ] 0 [ x1 (1 + ( x3 / x1 ) ] 0 Fig-3 presents the practical experiment on estimating/updating the Cartesian position of the Object over time Tracing using an EKF Estimated Measurement Y - Position X - Position Figure 3: Tracing the position of the target using a 1 st order EKF 3.3 Standard Particle Filter situation PF is an available method for tracing a variable of interest with (typically) a non- Gaussian and (potentially) multi-modal pdf [3]. The basis of this method is to use a sample-based construction to represent the entire pdf. Multiple samples (copies) of the variable of interest, i.e. the state of the process, are used in the PF, each one with the associated weight that signifies the quality of that sample; these are nown as particles. Instituto de Sistemas de Robótica - ISR 12 of 17

13 A formal (general) description of the PF algorithm is presented above Require: A set of particles at time 0: S 0 i =[x j,w j :j=1 M] W = w j :j=1 M {number of particles} while (Exploring) do = +1; if (ESS(W)<B*M) then {Particle population depleted} Index = Resample(W); S i = S i (index); end if for (j=1 to M) do {Prediction} x +1 j = f(x j,); end for S = Sensing(); for (j=1 to M) do {Update the weights} w +1 j = w j *W(s, x +1 j ); end for for (j=1 to M) do {Normalize the weights} w +1 j = w j /Σ w j end for end while Fig-4 presents the practical experiment on estimating/updating the Cartesian position of the Object over time. Figure 4: Tracing the position of the target using a Particle Filter Instituto de Sistemas de Robótica - ISR 13 of 17

14 3.4 PDAF method The Probabilistic Data Association (PDA) algorithm calculates the association probabilities for each validated measurement at the current time to the target of interest. This probabilistic (Bayesian) information is used in the tracing filter called PDA filter (PDAF), which accounts for the measurement origin uncertainty. This filter is particularly useful in the situation where a tracing filter receives more then one validated measurements, i.e. the origin uncertainty case: - From which target, if any, a particular measurement originated? In particular, the problem of extraneous measurements (due to random phenomena: false alarms or clutter) as well as neighbor targets. A measurement in the gate (i.e. a validated measurement), while not guaranteed to have originated from the target the gate pertains to, is a valid association candidate thus the name validation region or association region. If there is more then one detection (measurement) in the gate, this leads to an association uncertainty. The PDAF [2] algorithm is quite similar as in the KF based one. The Prediction (cycle) of the state and measurement is PREDICTION xˆ( + 1 ) = F( ) xˆ( ) zˆ( + 1 ) = H ( + 1) xˆ( + 1 ) (17) with covariance of the predicted state P ( + 1 ) = F( ) P( ) F( )' + Q( ) (18) Innovation covariance corresponding to the true measurement S ( + 1) = H ( + 1) P( + 1 ) H ( + 1)' + R( + 1) (19) MEASUREMENT SELECTION FOR UPDATE The validation region (gate/ellipsoid)* 1 υ( + 1) = { z :[ z zˆ( + 1 )]' S( + 1) [ z zˆ( + 1 )] λ} (20) where λ is the gate threshold determined by the chosen gate probability P G. *_ The ellipse (or ellipsoid) of probability concentration - the semi axis of the ellipse are the square roots of the eigenvalues of λs, where λ is the gate threshold and S is the covariance matrix of the innovation. Validated measurements { z i ( + 1), i = 1,..., m( + 1)} (21) Innovation corresponding to the i-th validated measurement υ ( + 1) = z ( + 1) zˆ( + 1 ) i = 1,..., m( + 1) (22) i i Instituto de Sistemas de Robótica - ISR 14 of 17

15 PROBABILISTIC DATA ASSOCIATION NONPARAMETRIC VERSION Probability that the i-th validated measurement is the correct one ei m( b + j βi( + 1) = b m( b + j + 1) = 1 + 1) = 1 e e j j i = 1,..., m( + 1) i = 0 (23) Where β 0 (+1) is the probability that none of the measurement is correct, and ei e n π b λ 1 1 i υi ( + 1)' S ( + 1) 2 2 z / 2 m( + 1) c 1 n υ ( + 1) z 1 PD P P D G (24) UPDATE State update xˆ ( ) = xˆ( + 1 ) + W ( + 1) ν ( + 1) (25) with the combined innovation m( + 1) i i= 1 ν ( + 1) β ( + 1) ν ( + 1) (26) Covariance associated with the updated state i ~ P( ) = P( + 1 ) [1 β 0 ( + 1)] W ( + 1) S( + 1) W ( + 1)' + P( + 1) (27) where the (weighted) spread of the innovation term is m( ~ ( + 1) ( + 1)[ + P W i= 1 1) β ( + 1) ν ( + 1) ν ( + 1)' ν ( + 1) ν ( + 1)'] W ( + 1)' (28) i i i A commented version of this algorithm is available on section (Annex) 5.2. Fig-5 presents the practical experiment on estimating/updating the Cartesian position of the Object over time. Instituto de Sistemas de Robótica - ISR 15 of 17

16 1200 Tracing of a 2D motion using the PDAF Estimated Detection Y-axis Position X-axis Position Figure 5: Tracing the position of the target using the PDAF Instituto de Sistemas de Robótica - ISR 16 of 17

17 4 References [1] Y. Bar-Shalom and T.E. Fortmann (1988). Tracing and Data Association. Academic Press. [2] Y. Bar-Shalom and X.R. Li (1995). Multitarget-Multisensor Tracing: Principles & Techniques. YBS Publishing. [3] Ioannis Releitis (2004). A Particle Filter Tutorial for Mobile Robot Localization. Technical Report TR-CIM-04-02, Centre for Intelligent Machines, McGill University, Montreal, Quebec, Canada. Instituto de Sistemas de Robótica - ISR 17 of 17

Robotics 2 Target Tracking. Kai Arras, Cyrill Stachniss, Maren Bennewitz, Wolfram Burgard

Robotics 2 Target Tracking. Kai Arras, Cyrill Stachniss, Maren Bennewitz, Wolfram Burgard Robotics 2 Target Tracking Kai Arras, Cyrill Stachniss, Maren Bennewitz, Wolfram Burgard Slides by Kai Arras, Gian Diego Tipaldi, v.1.1, Jan 2012 Chapter Contents Target Tracking Overview Applications

More information

1 Kalman Filter Introduction

1 Kalman Filter Introduction 1 Kalman Filter Introduction You should first read Chapter 1 of Stochastic models, estimation, and control: Volume 1 by Peter S. Maybec (available here). 1.1 Explanation of Equations (1-3) and (1-4) Equation

More information

A NEW FORMULATION OF IPDAF FOR TRACKING IN CLUTTER

A NEW FORMULATION OF IPDAF FOR TRACKING IN CLUTTER A NEW FRMULATIN F IPDAF FR TRACKING IN CLUTTER Jean Dezert NERA, 29 Av. Division Leclerc 92320 Châtillon, France fax:+33146734167 dezert@onera.fr Ning Li, X. Rong Li University of New rleans New rleans,

More information

Lecture Outline. Target Tracking: Lecture 3 Maneuvering Target Tracking Issues. Maneuver Illustration. Maneuver Illustration. Maneuver Detection

Lecture Outline. Target Tracking: Lecture 3 Maneuvering Target Tracking Issues. Maneuver Illustration. Maneuver Illustration. Maneuver Detection REGLERTEKNIK Lecture Outline AUTOMATIC CONTROL Target Tracking: Lecture 3 Maneuvering Target Tracking Issues Maneuver Detection Emre Özkan emre@isy.liu.se Division of Automatic Control Department of Electrical

More information

SLAM Techniques and Algorithms. Jack Collier. Canada. Recherche et développement pour la défense Canada. Defence Research and Development Canada

SLAM Techniques and Algorithms. Jack Collier. Canada. Recherche et développement pour la défense Canada. Defence Research and Development Canada SLAM Techniques and Algorithms Jack Collier Defence Research and Development Canada Recherche et développement pour la défense Canada Canada Goals What will we learn Gain an appreciation for what SLAM

More information

Previously on TT, Target Tracking: Lecture 2 Single Target Tracking Issues. Lecture-2 Outline. Basic ideas on track life

Previously on TT, Target Tracking: Lecture 2 Single Target Tracking Issues. Lecture-2 Outline. Basic ideas on track life REGLERTEKNIK Previously on TT, AUTOMATIC CONTROL Target Tracing: Lecture 2 Single Target Tracing Issues Emre Özan emre@isy.liu.se Division of Automatic Control Department of Electrical Engineering Linöping

More information

Bayes Filter Reminder. Kalman Filter Localization. Properties of Gaussians. Gaussians. Prediction. Correction. σ 2. Univariate. 1 2πσ e.

Bayes Filter Reminder. Kalman Filter Localization. Properties of Gaussians. Gaussians. Prediction. Correction. σ 2. Univariate. 1 2πσ e. Kalman Filter Localization Bayes Filter Reminder Prediction Correction Gaussians p(x) ~ N(µ,σ 2 ) : Properties of Gaussians Univariate p(x) = 1 1 2πσ e 2 (x µ) 2 σ 2 µ Univariate -σ σ Multivariate µ Multivariate

More information

State Estimation for Nonlinear Systems using Restricted Genetic Optimization

State Estimation for Nonlinear Systems using Restricted Genetic Optimization State Estimation for Nonlinear Systems using Restricted Genetic Optimization Santiago Garrido, Luis Moreno, and Carlos Balaguer Universidad Carlos III de Madrid, Leganés 28911, Madrid (Spain) Abstract.

More information

Tracking of Extended Objects and Group Targets using Random Matrices A New Approach

Tracking of Extended Objects and Group Targets using Random Matrices A New Approach Tracing of Extended Objects and Group Targets using Random Matrices A New Approach Michael Feldmann FGAN Research Institute for Communication, Information Processing and Ergonomics FKIE D-53343 Wachtberg,

More information

Robotics 2 Data Association. Giorgio Grisetti, Cyrill Stachniss, Kai Arras, Wolfram Burgard

Robotics 2 Data Association. Giorgio Grisetti, Cyrill Stachniss, Kai Arras, Wolfram Burgard Robotics 2 Data Association Giorgio Grisetti, Cyrill Stachniss, Kai Arras, Wolfram Burgard Data Association Data association is the process of associating uncertain measurements to known tracks. Problem

More information

EKF and SLAM. McGill COMP 765 Sept 18 th, 2017

EKF and SLAM. McGill COMP 765 Sept 18 th, 2017 EKF and SLAM McGill COMP 765 Sept 18 th, 2017 Outline News and information Instructions for paper presentations Continue on Kalman filter: EKF and extension to mapping Example of a real mapping system:

More information

ROBOTICS 01PEEQW. Basilio Bona DAUIN Politecnico di Torino

ROBOTICS 01PEEQW. Basilio Bona DAUIN Politecnico di Torino ROBOTICS 01PEEQW Basilio Bona DAUIN Politecnico di Torino Probabilistic Fundamentals in Robotics Gaussian Filters Course Outline Basic mathematical framework Probabilistic models of mobile robots Mobile

More information

Miscellaneous. Regarding reading materials. Again, ask questions (if you have) and ask them earlier

Miscellaneous. Regarding reading materials. Again, ask questions (if you have) and ask them earlier Miscellaneous Regarding reading materials Reading materials will be provided as needed If no assigned reading, it means I think the material from class is sufficient Should be enough for you to do your

More information

Robotics 2 Target Tracking. Giorgio Grisetti, Cyrill Stachniss, Kai Arras, Wolfram Burgard

Robotics 2 Target Tracking. Giorgio Grisetti, Cyrill Stachniss, Kai Arras, Wolfram Burgard Robotics 2 Target Tracking Giorgio Grisetti, Cyrill Stachniss, Kai Arras, Wolfram Burgard Linear Dynamical System (LDS) Stochastic process governed by is the state vector is the input vector is the process

More information

State Estimation by IMM Filter in the Presence of Structural Uncertainty 1

State Estimation by IMM Filter in the Presence of Structural Uncertainty 1 Recent Advances in Signal Processing and Communications Edited by Nios Mastorais World Scientific and Engineering Society (WSES) Press Greece 999 pp.8-88. State Estimation by IMM Filter in the Presence

More information

Kalman filtering and friends: Inference in time series models. Herke van Hoof slides mostly by Michael Rubinstein

Kalman filtering and friends: Inference in time series models. Herke van Hoof slides mostly by Michael Rubinstein Kalman filtering and friends: Inference in time series models Herke van Hoof slides mostly by Michael Rubinstein Problem overview Goal Estimate most probable state at time k using measurement up to time

More information

L06. LINEAR KALMAN FILTERS. NA568 Mobile Robotics: Methods & Algorithms

L06. LINEAR KALMAN FILTERS. NA568 Mobile Robotics: Methods & Algorithms L06. LINEAR KALMAN FILTERS NA568 Mobile Robotics: Methods & Algorithms 2 PS2 is out! Landmark-based Localization: EKF, UKF, PF Today s Lecture Minimum Mean Square Error (MMSE) Linear Kalman Filter Gaussian

More information

TARGET TRACKING AND DATA FUSION: How to Get the Most Out of Your Sensors and make a living out of it FUSION 2017

TARGET TRACKING AND DATA FUSION: How to Get the Most Out of Your Sensors and make a living out of it FUSION 2017 TARGET TRACKING AND DATA FUSION: How to Get the Most Out of Your Sensors and make a living out of it AN OVERVIEW OF TRACKING ALGORITHMS FOR CLUTTERED AND MULTITARGET-MULTISENSOR ENVIRONMENTS Yaakov Bar-Shalom,

More information

TSRT14: Sensor Fusion Lecture 9

TSRT14: Sensor Fusion Lecture 9 TSRT14: Sensor Fusion Lecture 9 Simultaneous localization and mapping (SLAM) Gustaf Hendeby gustaf.hendeby@liu.se TSRT14 Lecture 9 Gustaf Hendeby Spring 2018 1 / 28 Le 9: simultaneous localization and

More information

Nonlinear Parameter Estimation for State-Space ARCH Models with Missing Observations

Nonlinear Parameter Estimation for State-Space ARCH Models with Missing Observations Nonlinear Parameter Estimation for State-Space ARCH Models with Missing Observations SEBASTIÁN OSSANDÓN Pontificia Universidad Católica de Valparaíso Instituto de Matemáticas Blanco Viel 596, Cerro Barón,

More information

Data association uncertainty occurs when remote sensing devices, such as radar,

Data association uncertainty occurs when remote sensing devices, such as radar, The Probabilistic Data Association Filter ESTIMATION IN THE PRESENCE OF MEASUREMENT ORIGIN UNCERTAINTY YAAKOV BAR-SHALOM, FRED DAUM, and JIM HUANG Data association uncertainty occurs when remote sensing

More information

Autonomous Mobile Robot Design

Autonomous Mobile Robot Design Autonomous Mobile Robot Design Topic: Extended Kalman Filter Dr. Kostas Alexis (CSE) These slides relied on the lectures from C. Stachniss, J. Sturm and the book Probabilistic Robotics from Thurn et al.

More information

Kalman Filter. Predict: Update: x k k 1 = F k x k 1 k 1 + B k u k P k k 1 = F k P k 1 k 1 F T k + Q

Kalman Filter. Predict: Update: x k k 1 = F k x k 1 k 1 + B k u k P k k 1 = F k P k 1 k 1 F T k + Q Kalman Filter Kalman Filter Predict: x k k 1 = F k x k 1 k 1 + B k u k P k k 1 = F k P k 1 k 1 F T k + Q Update: K = P k k 1 Hk T (H k P k k 1 Hk T + R) 1 x k k = x k k 1 + K(z k H k x k k 1 ) P k k =(I

More information

Probabilistic Fundamentals in Robotics. DAUIN Politecnico di Torino July 2010

Probabilistic Fundamentals in Robotics. DAUIN Politecnico di Torino July 2010 Probabilistic Fundamentals in Robotics Gaussian Filters Basilio Bona DAUIN Politecnico di Torino July 2010 Course Outline Basic mathematical framework Probabilistic models of mobile robots Mobile robot

More information

From Bayes to Extended Kalman Filter

From Bayes to Extended Kalman Filter From Bayes to Extended Kalman Filter Michal Reinštein Czech Technical University in Prague Faculty of Electrical Engineering, Department of Cybernetics Center for Machine Perception http://cmp.felk.cvut.cz/

More information

ESTIMATOR STABILITY ANALYSIS IN SLAM. Teresa Vidal-Calleja, Juan Andrade-Cetto, Alberto Sanfeliu

ESTIMATOR STABILITY ANALYSIS IN SLAM. Teresa Vidal-Calleja, Juan Andrade-Cetto, Alberto Sanfeliu ESTIMATOR STABILITY ANALYSIS IN SLAM Teresa Vidal-Calleja, Juan Andrade-Cetto, Alberto Sanfeliu Institut de Robtica i Informtica Industrial, UPC-CSIC Llorens Artigas 4-6, Barcelona, 88 Spain {tvidal, cetto,

More information

Data Fusion Kalman Filtering Self Localization

Data Fusion Kalman Filtering Self Localization Data Fusion Kalman Filtering Self Localization Armando Jorge Sousa http://www.fe.up.pt/asousa asousa@fe.up.pt Faculty of Engineering, University of Porto, Portugal Department of Electrical and Computer

More information

STONY BROOK UNIVERSITY. CEAS Technical Report 829

STONY BROOK UNIVERSITY. CEAS Technical Report 829 1 STONY BROOK UNIVERSITY CEAS Technical Report 829 Variable and Multiple Target Tracking by Particle Filtering and Maximum Likelihood Monte Carlo Method Jaechan Lim January 4, 2006 2 Abstract In most applications

More information

Fisher Information Matrix-based Nonlinear System Conversion for State Estimation

Fisher Information Matrix-based Nonlinear System Conversion for State Estimation Fisher Information Matrix-based Nonlinear System Conversion for State Estimation Ming Lei Christophe Baehr and Pierre Del Moral Abstract In practical target tracing a number of improved measurement conversion

More information

PERFORMANCE COMPARISON OF TRACKING ALGORITHMS FOR A GROUND BASED RADAR GÖKHAN SOYSAL AND MURAT EFE

PERFORMANCE COMPARISON OF TRACKING ALGORITHMS FOR A GROUND BASED RADAR GÖKHAN SOYSAL AND MURAT EFE Commun. Fac. Sci. Univ. Ank. Series A2-A3 V.5() pp -6 (2007) PERFORMANCE COMPARISON OF TRACKING ALGORITHMS FOR A GROUND BASED RADAR Ankara University, Faculty of Engineering, Electronics Engineering Department,

More information

9 Multi-Model State Estimation

9 Multi-Model State Estimation Technion Israel Institute of Technology, Department of Electrical Engineering Estimation and Identification in Dynamical Systems (048825) Lecture Notes, Fall 2009, Prof. N. Shimkin 9 Multi-Model State

More information

Introduction p. 1 Fundamental Problems p. 2 Core of Fundamental Theory and General Mathematical Ideas p. 3 Classical Statistical Decision p.

Introduction p. 1 Fundamental Problems p. 2 Core of Fundamental Theory and General Mathematical Ideas p. 3 Classical Statistical Decision p. Preface p. xiii Acknowledgment p. xix Introduction p. 1 Fundamental Problems p. 2 Core of Fundamental Theory and General Mathematical Ideas p. 3 Classical Statistical Decision p. 4 Bayes Decision p. 5

More information

Tracking and Identification of Multiple targets

Tracking and Identification of Multiple targets Tracking and Identification of Multiple targets Samir Hachour, François Delmotte, Eric Lefèvre, David Mercier Laboratoire de Génie Informatique et d'automatique de l'artois, EA 3926 LGI2A first name.last

More information

Distributed estimation in sensor networks

Distributed estimation in sensor networks in sensor networks A. Benavoli Dpt. di Sistemi e Informatica Università di Firenze, Italy. e-mail: benavoli@dsi.unifi.it Outline 1 An introduction to 2 3 An introduction to An introduction to In recent

More information

Mini-Course 07 Kalman Particle Filters. Henrique Massard da Fonseca Cesar Cunha Pacheco Wellington Bettencurte Julio Dutra

Mini-Course 07 Kalman Particle Filters. Henrique Massard da Fonseca Cesar Cunha Pacheco Wellington Bettencurte Julio Dutra Mini-Course 07 Kalman Particle Filters Henrique Massard da Fonseca Cesar Cunha Pacheco Wellington Bettencurte Julio Dutra Agenda State Estimation Problems & Kalman Filter Henrique Massard Steady State

More information

Instituto de Sistemas e Robótica. Pólo de Lisboa

Instituto de Sistemas e Robótica. Pólo de Lisboa Instituto de Sistemas e Robótica Pólo de Lisboa Visual Tracking for Mobile Robot Localization 1 Jose Neira 2 October 1996 RT-602-96 ISR-Torre Norte Av. Rovisco Pais 1096 Lisboa CODEX PORTUGAL 1 This work

More information

2D Image Processing. Bayes filter implementation: Kalman filter

2D Image Processing. Bayes filter implementation: Kalman filter 2D Image Processing Bayes filter implementation: Kalman filter Prof. Didier Stricker Kaiserlautern University http://ags.cs.uni-kl.de/ DFKI Deutsches Forschungszentrum für Künstliche Intelligenz http://av.dfki.de

More information

Prediction of ESTSP Competition Time Series by Unscented Kalman Filter and RTS Smoother

Prediction of ESTSP Competition Time Series by Unscented Kalman Filter and RTS Smoother Prediction of ESTSP Competition Time Series by Unscented Kalman Filter and RTS Smoother Simo Särkkä, Aki Vehtari and Jouko Lampinen Helsinki University of Technology Department of Electrical and Communications

More information

A NOVEL OPTIMAL PROBABILITY DENSITY FUNCTION TRACKING FILTER DESIGN 1

A NOVEL OPTIMAL PROBABILITY DENSITY FUNCTION TRACKING FILTER DESIGN 1 A NOVEL OPTIMAL PROBABILITY DENSITY FUNCTION TRACKING FILTER DESIGN 1 Jinglin Zhou Hong Wang, Donghua Zhou Department of Automation, Tsinghua University, Beijing 100084, P. R. China Control Systems Centre,

More information

The Kernel-SME Filter with False and Missing Measurements

The Kernel-SME Filter with False and Missing Measurements The Kernel-SME Filter with False and Missing Measurements Marcus Baum, Shishan Yang Institute of Computer Science University of Göttingen, Germany Email: marcusbaum, shishanyang@csuni-goettingende Uwe

More information

Extended Object and Group Tracking: A Comparison of Random Matrices and Random Hypersurface Models

Extended Object and Group Tracking: A Comparison of Random Matrices and Random Hypersurface Models Extended Object and Group Tracking: A Comparison of Random Matrices and Random Hypersurface Models Marcus Baum, Michael Feldmann, Dietrich Fränken, Uwe D. Hanebeck, and Wolfgang Koch Intelligent Sensor-Actuator-Systems

More information

L11. EKF SLAM: PART I. NA568 Mobile Robotics: Methods & Algorithms

L11. EKF SLAM: PART I. NA568 Mobile Robotics: Methods & Algorithms L11. EKF SLAM: PART I NA568 Mobile Robotics: Methods & Algorithms Today s Topic EKF Feature-Based SLAM State Representation Process / Observation Models Landmark Initialization Robot-Landmark Correlation

More information

A Tree Search Approach to Target Tracking in Clutter

A Tree Search Approach to Target Tracking in Clutter 12th International Conference on Information Fusion Seattle, WA, USA, July 6-9, 2009 A Tree Search Approach to Target Tracking in Clutter Jill K. Nelson and Hossein Roufarshbaf Department of Electrical

More information

The Unscented Particle Filter

The Unscented Particle Filter The Unscented Particle Filter Rudolph van der Merwe (OGI) Nando de Freitas (UC Bereley) Arnaud Doucet (Cambridge University) Eric Wan (OGI) Outline Optimal Estimation & Filtering Optimal Recursive Bayesian

More information

Extended Object and Group Tracking with Elliptic Random Hypersurface Models

Extended Object and Group Tracking with Elliptic Random Hypersurface Models Extended Object and Group Tracing with Elliptic Random Hypersurface Models Marcus Baum Benjamin Noac and Uwe D. Hanebec Intelligent Sensor-Actuator-Systems Laboratory ISAS Institute for Anthropomatics

More information

Autonomous Navigation for Flying Robots

Autonomous Navigation for Flying Robots Computer Vision Group Prof. Daniel Cremers Autonomous Navigation for Flying Robots Lecture 6.2: Kalman Filter Jürgen Sturm Technische Universität München Motivation Bayes filter is a useful tool for state

More information

Combined Particle and Smooth Variable Structure Filtering for Nonlinear Estimation Problems

Combined Particle and Smooth Variable Structure Filtering for Nonlinear Estimation Problems 14th International Conference on Information Fusion Chicago, Illinois, USA, July 5-8, 2011 Combined Particle and Smooth Variable Structure Filtering for Nonlinear Estimation Problems S. Andrew Gadsden

More information

ADDRESSING TRACK COALESCENCE IN SEQUENTIAL K-BEST MULTIPLE HYPOTHESIS TRACKING

ADDRESSING TRACK COALESCENCE IN SEQUENTIAL K-BEST MULTIPLE HYPOTHESIS TRACKING ADDRESSING TRACK COALESCENCE IN SEQUENTIAL K-BEST MULTIPLE HYPOTHESIS TRACKING A Thesis Presented to The Academic Faculty By Ryan D. Palkki In Partial Fulfillment of the Requirements for the Degree Master

More information

Probabilistic Fundamentals in Robotics. DAUIN Politecnico di Torino July 2010

Probabilistic Fundamentals in Robotics. DAUIN Politecnico di Torino July 2010 Probabilistic Fundamentals in Robotics Probabilistic Models of Mobile Robots Robot localization Basilio Bona DAUIN Politecnico di Torino July 2010 Course Outline Basic mathematical framework Probabilistic

More information

GMTI Tracking in the Presence of Doppler and Range Ambiguities

GMTI Tracking in the Presence of Doppler and Range Ambiguities 14th International Conference on Information Fusion Chicago, Illinois, USA, July 5-8, 2011 GMTI Tracing in the Presence of Doppler and Range Ambiguities Michael Mertens Dept. Sensor Data and Information

More information

Minimum Necessary Data Rates for Accurate Track Fusion

Minimum Necessary Data Rates for Accurate Track Fusion Proceedings of the 44th IEEE Conference on Decision Control, the European Control Conference 005 Seville, Spain, December -5, 005 ThIA0.4 Minimum Necessary Data Rates for Accurate Trac Fusion Barbara F.

More information

A Study of Covariances within Basic and Extended Kalman Filters

A Study of Covariances within Basic and Extended Kalman Filters A Study of Covariances within Basic and Extended Kalman Filters David Wheeler Kyle Ingersoll December 2, 2013 Abstract This paper explores the role of covariance in the context of Kalman filters. The underlying

More information

Sensor Fusion: Particle Filter

Sensor Fusion: Particle Filter Sensor Fusion: Particle Filter By: Gordana Stojceska stojcesk@in.tum.de Outline Motivation Applications Fundamentals Tracking People Advantages and disadvantages Summary June 05 JASS '05, St.Petersburg,

More information

F denotes cumulative density. denotes probability density function; (.)

F denotes cumulative density. denotes probability density function; (.) BAYESIAN ANALYSIS: FOREWORDS Notation. System means the real thing and a model is an assumed mathematical form for the system.. he probability model class M contains the set of the all admissible models

More information

Self-Driving Car ND - Sensor Fusion - Extended Kalman Filters

Self-Driving Car ND - Sensor Fusion - Extended Kalman Filters Self-Driving Car ND - Sensor Fusion - Extended Kalman Filters Udacity and Mercedes February 7, 07 Introduction Lesson Ma 3 Estimation Problem Refresh 4 Measurement Udate Quiz 5 Kalman Filter Equations

More information

Conditions for Suboptimal Filter Stability in SLAM

Conditions for Suboptimal Filter Stability in SLAM Conditions for Suboptimal Filter Stability in SLAM Teresa Vidal-Calleja, Juan Andrade-Cetto and Alberto Sanfeliu Institut de Robòtica i Informàtica Industrial, UPC-CSIC Llorens Artigas -, Barcelona, Spain

More information

Optimal Mixture Approximation of the Product of Mixtures

Optimal Mixture Approximation of the Product of Mixtures Optimal Mixture Approximation of the Product of Mixtures Oliver C Schrempf, Olga Feiermann, and Uwe D Hanebeck Intelligent Sensor-Actuator-Systems Laboratory Institute of Computer Science and Engineering

More information

Conditional Posterior Cramér-Rao Lower Bounds for Nonlinear Sequential Bayesian Estimation

Conditional Posterior Cramér-Rao Lower Bounds for Nonlinear Sequential Bayesian Estimation 1 Conditional Posterior Cramér-Rao Lower Bounds for Nonlinear Sequential Bayesian Estimation Long Zuo, Ruixin Niu, and Pramod K. Varshney Abstract Posterior Cramér-Rao lower bounds (PCRLBs) 1] for sequential

More information

Summary of Past Lectures. Target Tracking: Lecture 4 Multiple Target Tracking: Part I. Lecture Outline. What is a hypothesis?

Summary of Past Lectures. Target Tracking: Lecture 4 Multiple Target Tracking: Part I. Lecture Outline. What is a hypothesis? REGLERTEKNIK Summary of Past Lectures AUTOMATIC COROL Target Tracing: Lecture Multiple Target Tracing: Part I Emre Özan emre@isy.liu.se Division of Automatic Control Department of Electrical Engineering

More information

Lecture Outline. Target Tracking: Lecture 7 Multiple Sensor Tracking Issues. Multi Sensor Architectures. Multi Sensor Architectures

Lecture Outline. Target Tracking: Lecture 7 Multiple Sensor Tracking Issues. Multi Sensor Architectures. Multi Sensor Architectures Lecture Outline Target Tracing: Lecture 7 Multiple Sensor Tracing Issues Umut Orguner umut@metu.edu.tr room: EZ-12 tel: 4425 Department of Electrical & Electronics Engineering Middle East Technical University

More information

Evaluating the Consistency of Estimation

Evaluating the Consistency of Estimation Tampere University of Technology Evaluating the Consistency of Estimation Citation Ivanov, P., Ali-Löytty, S., & Piche, R. (201). Evaluating the Consistency of Estimation. In Proceedings of 201 International

More information

2D Image Processing. Bayes filter implementation: Kalman filter

2D Image Processing. Bayes filter implementation: Kalman filter 2D Image Processing Bayes filter implementation: Kalman filter Prof. Didier Stricker Dr. Gabriele Bleser Kaiserlautern University http://ags.cs.uni-kl.de/ DFKI Deutsches Forschungszentrum für Künstliche

More information

Simultaneous Localization and Mapping (SLAM) Corso di Robotica Prof. Davide Brugali Università degli Studi di Bergamo

Simultaneous Localization and Mapping (SLAM) Corso di Robotica Prof. Davide Brugali Università degli Studi di Bergamo Simultaneous Localization and Mapping (SLAM) Corso di Robotica Prof. Davide Brugali Università degli Studi di Bergamo Introduction SLAM asks the following question: Is it possible for an autonomous vehicle

More information

Gaussian Mixtures Proposal Density in Particle Filter for Track-Before-Detect

Gaussian Mixtures Proposal Density in Particle Filter for Track-Before-Detect 12th International Conference on Information Fusion Seattle, WA, USA, July 6-9, 29 Gaussian Mixtures Proposal Density in Particle Filter for Trac-Before-Detect Ondřej Straa, Miroslav Šimandl and Jindřich

More information

Systematic Error Modeling and Bias Estimation

Systematic Error Modeling and Bias Estimation sensors Article Systematic Error Modeling and Bias Estimation Feihu Zhang * and Alois Knoll Robotics and Embedded Systems, Technische Universität München, 8333 München, Germany; knoll@in.tum.de * Correspondence:

More information

EKF, UKF. Pieter Abbeel UC Berkeley EECS. Many slides adapted from Thrun, Burgard and Fox, Probabilistic Robotics

EKF, UKF. Pieter Abbeel UC Berkeley EECS. Many slides adapted from Thrun, Burgard and Fox, Probabilistic Robotics EKF, UKF Pieter Abbeel UC Berkeley EECS Many slides adapted from Thrun, Burgard and Fox, Probabilistic Robotics Kalman Filter Kalman Filter = special case of a Bayes filter with dynamics model and sensory

More information

EKF, UKF. Pieter Abbeel UC Berkeley EECS. Many slides adapted from Thrun, Burgard and Fox, Probabilistic Robotics

EKF, UKF. Pieter Abbeel UC Berkeley EECS. Many slides adapted from Thrun, Burgard and Fox, Probabilistic Robotics EKF, UKF Pieter Abbeel UC Berkeley EECS Many slides adapted from Thrun, Burgard and Fox, Probabilistic Robotics Kalman Filter Kalman Filter = special case of a Bayes filter with dynamics model and sensory

More information

in a Rao-Blackwellised Unscented Kalman Filter

in a Rao-Blackwellised Unscented Kalman Filter A Rao-Blacwellised Unscented Kalman Filter Mar Briers QinetiQ Ltd. Malvern Technology Centre Malvern, UK. m.briers@signal.qinetiq.com Simon R. Masell QinetiQ Ltd. Malvern Technology Centre Malvern, UK.

More information

A Gaussian Mixture PHD Filter for Nonlinear Jump Markov Models

A Gaussian Mixture PHD Filter for Nonlinear Jump Markov Models A Gaussian Mixture PHD Filter for Nonlinear Jump Marov Models Ba-Ngu Vo Ahmed Pasha Hoang Duong Tuan Department of Electrical and Electronic Engineering The University of Melbourne Parville VIC 35 Australia

More information

Robot Localization and Kalman Filters

Robot Localization and Kalman Filters Robot Localization and Kalman Filters Rudy Negenborn rudy@negenborn.net August 26, 2003 Outline Robot Localization Probabilistic Localization Kalman Filters Kalman Localization Kalman Localization with

More information

CS 532: 3D Computer Vision 6 th Set of Notes

CS 532: 3D Computer Vision 6 th Set of Notes 1 CS 532: 3D Computer Vision 6 th Set of Notes Instructor: Philippos Mordohai Webpage: www.cs.stevens.edu/~mordohai E-mail: Philippos.Mordohai@stevens.edu Office: Lieb 215 Lecture Outline Intro to Covariance

More information

2D Image Processing (Extended) Kalman and particle filter

2D Image Processing (Extended) Kalman and particle filter 2D Image Processing (Extended) Kalman and particle filter Prof. Didier Stricker Dr. Gabriele Bleser Kaiserlautern University http://ags.cs.uni-kl.de/ DFKI Deutsches Forschungszentrum für Künstliche Intelligenz

More information

Introduction to Unscented Kalman Filter

Introduction to Unscented Kalman Filter Introduction to Unscented Kalman Filter 1 Introdution In many scientific fields, we use certain models to describe the dynamics of system, such as mobile robot, vision tracking and so on. The word dynamics

More information

MMSE-Based Filtering for Linear and Nonlinear Systems in the Presence of Non-Gaussian System and Measurement Noise

MMSE-Based Filtering for Linear and Nonlinear Systems in the Presence of Non-Gaussian System and Measurement Noise MMSE-Based Filtering for Linear and Nonlinear Systems in the Presence of Non-Gaussian System and Measurement Noise I. Bilik 1 and J. Tabrikian 2 1 Dept. of Electrical and Computer Engineering, University

More information

Probabilistic Fundamentals in Robotics

Probabilistic Fundamentals in Robotics Probabilistic Fundamentals in Robotics Probabilistic Models of Mobile Robots Robot localization Basilio Bona DAUIN Politecnico di Torino June 2011 Course Outline Basic mathematical framework Probabilistic

More information

Perception: objects in the environment

Perception: objects in the environment Zsolt Vizi, Ph.D. 2018 Self-driving cars Sensor fusion: one categorization Type 1: low-level/raw data fusion combining several sources of raw data to produce new data that is expected to be more informative

More information

Cardinality Balanced Multi-Target Multi-Bernoulli Filtering Using Adaptive Birth Distributions

Cardinality Balanced Multi-Target Multi-Bernoulli Filtering Using Adaptive Birth Distributions Cardinality Balanced Multi-Target Multi-Bernoulli Filtering Using Adaptive Birth Distributions Stephan Reuter, Daniel Meissner, Benjamin Wiling, and Klaus Dietmayer Institute of Measurement, Control, and

More information

NON-LINEAR NOISE ADAPTIVE KALMAN FILTERING VIA VARIATIONAL BAYES

NON-LINEAR NOISE ADAPTIVE KALMAN FILTERING VIA VARIATIONAL BAYES 2013 IEEE INTERNATIONAL WORKSHOP ON MACHINE LEARNING FOR SIGNAL PROCESSING NON-LINEAR NOISE ADAPTIVE KALMAN FILTERING VIA VARIATIONAL BAYES Simo Särä Aalto University, 02150 Espoo, Finland Jouni Hartiainen

More information

Constrained State Estimation Using the Unscented Kalman Filter

Constrained State Estimation Using the Unscented Kalman Filter 16th Mediterranean Conference on Control and Automation Congress Centre, Ajaccio, France June 25-27, 28 Constrained State Estimation Using the Unscented Kalman Filter Rambabu Kandepu, Lars Imsland and

More information

Sequential Bayesian Estimation of the Probability of Detection for Tracking

Sequential Bayesian Estimation of the Probability of Detection for Tracking 2th International Conference on Information Fusion Seattle, WA, USA, July 6-9, 2009 Sequential Bayesian Estimation of the Probability of Detection for Tracking Kevin G. Jamieson Applied Physics Lab University

More information

Target Tracking and Classification using Collaborative Sensor Networks

Target Tracking and Classification using Collaborative Sensor Networks Target Tracking and Classification using Collaborative Sensor Networks Xiaodong Wang Department of Electrical Engineering Columbia University p.1/3 Talk Outline Background on distributed wireless sensor

More information

A Gaussian Mixture Motion Model and Contact Fusion Applied to the Metron Data Set

A Gaussian Mixture Motion Model and Contact Fusion Applied to the Metron Data Set 1th International Conference on Information Fusion Chicago, Illinois, USA, July 5-8, 211 A Gaussian Mixture Motion Model and Contact Fusion Applied to the Metron Data Set Kathrin Wilkens 1,2 1 Institute

More information

Nonlinear Estimation Techniques for Impact Point Prediction of Ballistic Targets

Nonlinear Estimation Techniques for Impact Point Prediction of Ballistic Targets Nonlinear Estimation Techniques for Impact Point Prediction of Ballistic Targets J. Clayton Kerce a, George C. Brown a, and David F. Hardiman b a Georgia Tech Research Institute, Georgia Institute of Technology,

More information

Rao-Blackwellized Particle Filter for Multiple Target Tracking

Rao-Blackwellized Particle Filter for Multiple Target Tracking Rao-Blackwellized Particle Filter for Multiple Target Tracking Simo Särkkä, Aki Vehtari, Jouko Lampinen Helsinki University of Technology, Finland Abstract In this article we propose a new Rao-Blackwellized

More information

Using the Kalman Filter to Estimate the State of a Maneuvering Aircraft

Using the Kalman Filter to Estimate the State of a Maneuvering Aircraft 1 Using the Kalman Filter to Estimate the State of a Maneuvering Aircraft K. Meier and A. Desai Abstract Using sensors that only measure the bearing angle and range of an aircraft, a Kalman filter is implemented

More information

Distributed Data Fusion with Kalman Filters. Simon Julier Computer Science Department University College London

Distributed Data Fusion with Kalman Filters. Simon Julier Computer Science Department University College London Distributed Data Fusion with Kalman Filters Simon Julier Computer Science Department University College London S.Julier@cs.ucl.ac.uk Structure of Talk Motivation Kalman Filters Double Counting Optimal

More information

A Comparison of Particle Filters for Personal Positioning

A Comparison of Particle Filters for Personal Positioning VI Hotine-Marussi Symposium of Theoretical and Computational Geodesy May 9-June 6. A Comparison of Particle Filters for Personal Positioning D. Petrovich and R. Piché Institute of Mathematics Tampere University

More information

A New Nonlinear Filtering Method for Ballistic Target Tracking

A New Nonlinear Filtering Method for Ballistic Target Tracking th International Conference on Information Fusion Seattle, WA, USA, July 6-9, 9 A New Nonlinear Filtering Method for Ballistic arget racing Chunling Wu Institute of Electronic & Information Engineering

More information

Sensor Tasking and Control

Sensor Tasking and Control Sensor Tasking and Control Sensing Networking Leonidas Guibas Stanford University Computation CS428 Sensor systems are about sensing, after all... System State Continuous and Discrete Variables The quantities

More information

TSRT14: Sensor Fusion Lecture 8

TSRT14: Sensor Fusion Lecture 8 TSRT14: Sensor Fusion Lecture 8 Particle filter theory Marginalized particle filter Gustaf Hendeby gustaf.hendeby@liu.se TSRT14 Lecture 8 Gustaf Hendeby Spring 2018 1 / 25 Le 8: particle filter theory,

More information

A Matrix Theoretic Derivation of the Kalman Filter

A Matrix Theoretic Derivation of the Kalman Filter A Matrix Theoretic Derivation of the Kalman Filter 4 September 2008 Abstract This paper presents a matrix-theoretic derivation of the Kalman filter that is accessible to students with a strong grounding

More information

Probability Map Building of Uncertain Dynamic Environments with Indistinguishable Obstacles

Probability Map Building of Uncertain Dynamic Environments with Indistinguishable Obstacles Probability Map Building of Uncertain Dynamic Environments with Indistinguishable Obstacles Myungsoo Jun and Raffaello D Andrea Sibley School of Mechanical and Aerospace Engineering Cornell University

More information

Censoring and Fusion in Non-linear Distributed Tracking Systems with Application to 2D Radar

Censoring and Fusion in Non-linear Distributed Tracking Systems with Application to 2D Radar Virginia Commonwealth University VCU Scholars Compass Theses and Dissertations Graduate School 15 Censoring and Fusion in Non-linear Distributed Tracking Systems with Application to D Radar Armond S. Conte

More information

Lego NXT: Navigation and localization using infrared distance sensors and Extended Kalman Filter. Miguel Pinto, A. Paulo Moreira, Aníbal Matos

Lego NXT: Navigation and localization using infrared distance sensors and Extended Kalman Filter. Miguel Pinto, A. Paulo Moreira, Aníbal Matos Lego NXT: Navigation and localization using infrared distance sensors and Extended Kalman Filter Miguel Pinto, A. Paulo Moreira, Aníbal Matos 1 Resume LegoFeup Localization Real and simulated scenarios

More information

Random Finite Set Methods. for Multitarget Tracking

Random Finite Set Methods. for Multitarget Tracking Random Finite Set Methods for Multitarget Tracing RANDOM FINITE SET METHODS FOR MULTITARGET TRACKING BY DARCY DUNNE a thesis submitted to the department of electrical & computer engineering and the school

More information

the robot in its current estimated position and orientation (also include a point at the reference point of the robot)

the robot in its current estimated position and orientation (also include a point at the reference point of the robot) CSCI 4190 Introduction to Robotic Algorithms, Spring 006 Assignment : out February 13, due February 3 and March Localization and the extended Kalman filter In this assignment, you will write a program

More information

Gaussian Processes. Le Song. Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012

Gaussian Processes. Le Song. Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012 Gaussian Processes Le Song Machine Learning II: Advanced Topics CSE 8803ML, Spring 01 Pictorial view of embedding distribution Transform the entire distribution to expected features Feature space Feature

More information

Data assimilation with and without a model

Data assimilation with and without a model Data assimilation with and without a model Tim Sauer George Mason University Parameter estimation and UQ U. Pittsburgh Mar. 5, 2017 Partially supported by NSF Most of this work is due to: Tyrus Berry,

More information

Lecture 2: From Linear Regression to Kalman Filter and Beyond

Lecture 2: From Linear Regression to Kalman Filter and Beyond Lecture 2: From Linear Regression to Kalman Filter and Beyond Department of Biomedical Engineering and Computational Science Aalto University January 26, 2012 Contents 1 Batch and Recursive Estimation

More information

Tuning of Extended Kalman Filter for nonlinear State Estimation

Tuning of Extended Kalman Filter for nonlinear State Estimation OSR Journal of Computer Engineering (OSR-JCE) e-ssn: 78-0661,p-SSN: 78-877, Volume 18, ssue 5, Ver. V (Sep. - Oct. 016), PP 14-19 www.iosrjournals.org Tuning of Extended Kalman Filter for nonlinear State

More information