Signal Processing, Sensor/Information Fusion, and Target Recognition XXIX, 2020
This paper is generally concerned with mathematical formalisms to support theory and algorithm de... more This paper is generally concerned with mathematical formalisms to support theory and algorithm developments of multiple hypothesis tracking (MHT), as a class of solutions to multiple target tracking (MTT) problems based on targetwise detections. In particular, this paper presents a new perspective on random set (RFSet) formalism to support a form of MHT, in which an unknown number of targets is modeled by a RFSet of continuous-time stochastic processes, rather than a single stochastic process defined on the space of finite sets in a given target state space, while generally multiple sensors provide noisy and cluttered target detections without any explicit indications of origins. The focus is on a clearcut approach to avoid any complication resulting from diagonal sets in direct-product spaces when a space of finite subsets of a state space is defined as its quotient space, instead of a subspace of the space of closed subsets in the state space with Fell-Matheron topology.
2018 21st International Conference on Information Fusion (FUSION), 2018
This paper reviews forty years of distributed estimation research since the first papers on decen... more This paper reviews forty years of distributed estimation research since the first papers on decentralized filtering appeared in 1978. Starting with a formulation of the problem, it reviews the assumptions and objectives of the main approaches, including information decorrelation, cross-covariance fusion, channel filters, covariance intersection, maximum a posteriori probability fusion, best linear unbiased estimate, and distributed Kalman filters based on pseudo estimates and augmented state estimates. It also reviews algorithms motivated by sensor networks with flexible communication including consensus and diffusion filters. Suggestions for future research are provided.
Based on bargaining problems in the form that J. Nash originally formulated, a two-person bargain... more Based on bargaining problems in the form that J. Nash originally formulated, a two-person bargaining process is modelled as an infinite game in extensive form. In this game, both bargainers make their proposals in terms of the utility pair resulting from possible agreement, and express approval or disapproval, alternatively changing roles. The existence and the uniqueness of subgame-perfect pure-strategy Nash equilibrium are explored. Limiting results when the time between bargaining sessions becomes shorter and shorter are connected with one of the conventional bargaining theories. Throughout this note, time-preference modelled by constant discounting plays an important role.
2018 21st International Conference on Information Fusion (FUSION), 2018
Multiple hypothesis tracking addresses difficult multiple target tracking problems by making asso... more Multiple hypothesis tracking addresses difficult multiple target tracking problems by making association decisions using multiple scans or frames of data. This paper reviews forty years of its development, including the original measurement-oriented approach of Reid, track-oriented approach first formulated by Morefield, distributed processing, and recent graph-based approaches. It also discusses its relationship with random set approaches for tracking.
This paper describes a continuous-time-state-process, discrete-time-observation, Interacting Mult... more This paper describes a continuous-time-state-process, discrete-time-observation, Interacting Multiple Model (IMM) tracking algorithm, and its applications to financial market modeling and asset allocation. A system state is modeled as a continuous-time, affine-Gaussian stochastic dynamical process driven by a white process noise, as well as structural changes modeled by a finite-state, continuous-time, Markov process. The system generally assumes multiple models with different state space dimensions and an affine-Gaussian state jump whenever a model transition occurs. The underlying problem is a standard filtering problem for estimating the system state based on a sequence of discrete-time, linear-Gaussian observations of partial system states. As our first attempt for applying the IMM methods to financial market modeling, we will use a rather naïve switching process using simple multiple linear stochastic system models.
In this paper, we consider a general twosensor, track-to-track association problem, in which an u... more In this paper, we consider a general twosensor, track-to-track association problem, in which an unknown number of targets is modeled as an independent, identically distributed (i.i.d.) system of random elements in a given target state space, while the a priori probability distribution of the total number of targets is not necessarily Poisson. We will show that, in order to accommodate not-necessarily-Poisson distributions, we need to modify the well-known, commonly-used track-to-track association hypothesis evaluation formula, by adding an extra multiplier that is a function of the hypothesized number of the detected targets. In order for this multiplier to be constant, thereby allowing us to use the commonly used track association metric, the Poisson assumption is not only sufficient but also necessary. A general multiple target tracking problem is a dynamic state estimation (filtering) problem in which the system state is that of a set of an unknown number of objects and the obser...
The multi-target tracking problem is both theoret-ically interesting and very important in terms ... more The multi-target tracking problem is both theoret-ically interesting and very important in terms of applications. Past results have been well documented in the survey paper [1] and theNaval Ocean Surveil-lance CorrelationHandbooks, [2] and [3]. The intro-ductory section of [4] ...
Abstract The tracking and classification of multiple targets by a network of local agents (nodes)... more Abstract The tracking and classification of multiple targets by a network of local agents (nodes) is considered. A Bayesian approach is adopted as the theoretical basis. Each local agent processes the local sensor data to obtain the local information state consisting of ...
This paper considers distributed multitarget track-ing in a distributed sensor network(DSN). A DS... more This paper considers distributed multitarget track-ing in a distributed sensor network(DSN). A DSN is made up of a set of nodes which can communicate to each other via a communication network (Figure 1). Each DSN node contains a processor collecting data from some ...
In this paper, we consider the distributed estima-tion problem by a set of agents connected by an... more In this paper, we consider the distributed estima-tion problem by a set of agents connected by an srbi-trary comunication network. Specifically, the problem of reconstructing the conditional probability of the random state using the conditional probabilities com-municated from ...
ABSTRACT This report presents research results on distributed situation assessment in a distribut... more ABSTRACT This report presents research results on distributed situation assessment in a distributed sensor network (DSN). The area of multitarget tracking and classification has been chosen to investigate issues associated with distributed hypothesis formation and evaluation. A general theory for Bayesian multitarget tracking has been developed. This is used as the basis for specifying the processing architecture at each node in the DSN. Each node contains the Generalized Tracker/Classifier for processing of local sensor data, an information fusion module to integrate processed information from various nodes, and an information distribution module. The problem of removing redundant information in a general distributed estimation system has also been investigated. Simulation results to study various issues associated with distributed situation assessment are presented. (Author)
This paper discusses distributed multitarget track-ing in a distributed sensor network (DSN). Eac... more This paper discusses distributed multitarget track-ing in a distributed sensor network (DSN). Each node in the network performs tracking functions using the local sensor data and communicates the processing results to other nodes according to some communication strategy. The ...
2009 12th International Conference on Information Fusion, Jul 6, 2009
... Chee-Yee Chong, Greg Castanon, Nathan Cooprider Shozo Mori, Ravi Ravichandran BAE Systems Bur... more ... Chee-Yee Chong, Greg Castanon, Nathan Cooprider Shozo Mori, Ravi Ravichandran BAE Systems Burlington, MA, USA {chee.chong, greg.castanon, nathan.cooprider, shozo.mori balasubramaniam.ravichandran}@baesystems.com ... ( )k j τ = states that the j-th report in the k-...
Signal Processing, Sensor/Information Fusion, and Target Recognition XXIX, 2020
This paper is generally concerned with mathematical formalisms to support theory and algorithm de... more This paper is generally concerned with mathematical formalisms to support theory and algorithm developments of multiple hypothesis tracking (MHT), as a class of solutions to multiple target tracking (MTT) problems based on targetwise detections. In particular, this paper presents a new perspective on random set (RFSet) formalism to support a form of MHT, in which an unknown number of targets is modeled by a RFSet of continuous-time stochastic processes, rather than a single stochastic process defined on the space of finite sets in a given target state space, while generally multiple sensors provide noisy and cluttered target detections without any explicit indications of origins. The focus is on a clearcut approach to avoid any complication resulting from diagonal sets in direct-product spaces when a space of finite subsets of a state space is defined as its quotient space, instead of a subspace of the space of closed subsets in the state space with Fell-Matheron topology.
2018 21st International Conference on Information Fusion (FUSION), 2018
This paper reviews forty years of distributed estimation research since the first papers on decen... more This paper reviews forty years of distributed estimation research since the first papers on decentralized filtering appeared in 1978. Starting with a formulation of the problem, it reviews the assumptions and objectives of the main approaches, including information decorrelation, cross-covariance fusion, channel filters, covariance intersection, maximum a posteriori probability fusion, best linear unbiased estimate, and distributed Kalman filters based on pseudo estimates and augmented state estimates. It also reviews algorithms motivated by sensor networks with flexible communication including consensus and diffusion filters. Suggestions for future research are provided.
Based on bargaining problems in the form that J. Nash originally formulated, a two-person bargain... more Based on bargaining problems in the form that J. Nash originally formulated, a two-person bargaining process is modelled as an infinite game in extensive form. In this game, both bargainers make their proposals in terms of the utility pair resulting from possible agreement, and express approval or disapproval, alternatively changing roles. The existence and the uniqueness of subgame-perfect pure-strategy Nash equilibrium are explored. Limiting results when the time between bargaining sessions becomes shorter and shorter are connected with one of the conventional bargaining theories. Throughout this note, time-preference modelled by constant discounting plays an important role.
2018 21st International Conference on Information Fusion (FUSION), 2018
Multiple hypothesis tracking addresses difficult multiple target tracking problems by making asso... more Multiple hypothesis tracking addresses difficult multiple target tracking problems by making association decisions using multiple scans or frames of data. This paper reviews forty years of its development, including the original measurement-oriented approach of Reid, track-oriented approach first formulated by Morefield, distributed processing, and recent graph-based approaches. It also discusses its relationship with random set approaches for tracking.
This paper describes a continuous-time-state-process, discrete-time-observation, Interacting Mult... more This paper describes a continuous-time-state-process, discrete-time-observation, Interacting Multiple Model (IMM) tracking algorithm, and its applications to financial market modeling and asset allocation. A system state is modeled as a continuous-time, affine-Gaussian stochastic dynamical process driven by a white process noise, as well as structural changes modeled by a finite-state, continuous-time, Markov process. The system generally assumes multiple models with different state space dimensions and an affine-Gaussian state jump whenever a model transition occurs. The underlying problem is a standard filtering problem for estimating the system state based on a sequence of discrete-time, linear-Gaussian observations of partial system states. As our first attempt for applying the IMM methods to financial market modeling, we will use a rather naïve switching process using simple multiple linear stochastic system models.
In this paper, we consider a general twosensor, track-to-track association problem, in which an u... more In this paper, we consider a general twosensor, track-to-track association problem, in which an unknown number of targets is modeled as an independent, identically distributed (i.i.d.) system of random elements in a given target state space, while the a priori probability distribution of the total number of targets is not necessarily Poisson. We will show that, in order to accommodate not-necessarily-Poisson distributions, we need to modify the well-known, commonly-used track-to-track association hypothesis evaluation formula, by adding an extra multiplier that is a function of the hypothesized number of the detected targets. In order for this multiplier to be constant, thereby allowing us to use the commonly used track association metric, the Poisson assumption is not only sufficient but also necessary. A general multiple target tracking problem is a dynamic state estimation (filtering) problem in which the system state is that of a set of an unknown number of objects and the obser...
The multi-target tracking problem is both theoret-ically interesting and very important in terms ... more The multi-target tracking problem is both theoret-ically interesting and very important in terms of applications. Past results have been well documented in the survey paper [1] and theNaval Ocean Surveil-lance CorrelationHandbooks, [2] and [3]. The intro-ductory section of [4] ...
Abstract The tracking and classification of multiple targets by a network of local agents (nodes)... more Abstract The tracking and classification of multiple targets by a network of local agents (nodes) is considered. A Bayesian approach is adopted as the theoretical basis. Each local agent processes the local sensor data to obtain the local information state consisting of ...
This paper considers distributed multitarget track-ing in a distributed sensor network(DSN). A DS... more This paper considers distributed multitarget track-ing in a distributed sensor network(DSN). A DSN is made up of a set of nodes which can communicate to each other via a communication network (Figure 1). Each DSN node contains a processor collecting data from some ...
In this paper, we consider the distributed estima-tion problem by a set of agents connected by an... more In this paper, we consider the distributed estima-tion problem by a set of agents connected by an srbi-trary comunication network. Specifically, the problem of reconstructing the conditional probability of the random state using the conditional probabilities com-municated from ...
ABSTRACT This report presents research results on distributed situation assessment in a distribut... more ABSTRACT This report presents research results on distributed situation assessment in a distributed sensor network (DSN). The area of multitarget tracking and classification has been chosen to investigate issues associated with distributed hypothesis formation and evaluation. A general theory for Bayesian multitarget tracking has been developed. This is used as the basis for specifying the processing architecture at each node in the DSN. Each node contains the Generalized Tracker/Classifier for processing of local sensor data, an information fusion module to integrate processed information from various nodes, and an information distribution module. The problem of removing redundant information in a general distributed estimation system has also been investigated. Simulation results to study various issues associated with distributed situation assessment are presented. (Author)
This paper discusses distributed multitarget track-ing in a distributed sensor network (DSN). Eac... more This paper discusses distributed multitarget track-ing in a distributed sensor network (DSN). Each node in the network performs tracking functions using the local sensor data and communicates the processing results to other nodes according to some communication strategy. The ...
2009 12th International Conference on Information Fusion, Jul 6, 2009
... Chee-Yee Chong, Greg Castanon, Nathan Cooprider Shozo Mori, Ravi Ravichandran BAE Systems Bur... more ... Chee-Yee Chong, Greg Castanon, Nathan Cooprider Shozo Mori, Ravi Ravichandran BAE Systems Burlington, MA, USA {chee.chong, greg.castanon, nathan.cooprider, shozo.mori balasubramaniam.ravichandran}@baesystems.com ... ( )k j τ = states that the j-th report in the k-...
Uploads
Papers