Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


Dissertation.pdf (18.41 MB)

ETD Abstract Container

Abstract Header

On Computationally Efficient Frameworks For Data Association In Multi-Target Tracking

Krishnaswamy, Sriram
http://rave.ohiolink.edu/etdc/view?acc_num=osu1574672274983947

Abstract Details

2019, Doctor of Philosophy, Ohio State University, Mechanical Engineering.
The aim of this dissertation is to examine ways of improving the computational efficiency of data association algorithms in tracking and to do so with better methods to handle data. Data association algorithms are employed in tracking problems in conjunction with an estimation algorithm to determine the optimal state estimate of multiple objects of interest given a set of measurements. This work primarily deals with Bayesian or pseudo-Bayesian paradigms for data association and reduces the computational cost by reducing the exponential growth or the so-called ``curse of dimensionality'' in these problems. This increase in the number of hypotheses is exacerbated in dense environments with low signal-to-noise ratio (SNR). This research employs tensor decomposition to reduce the number of incoming measurements into a core tensor or a low-dimensional summary and use it as a substitute for the complete set of measurements. The underlying data association considered in this research is the Joint Probabilistic Data Association (JPDA), a pseudo-Bayesian sub-optimal single-scan data association algorithm for multiple measurements. JPDA determines the best measurement by constructing a series of hypotheses, known as feasible events, based on a binary matrix, known as validation matrix, that indicates if a given measurement could have originated from a target. The primary bottleneck in JPDA is that the number of feasible events grows exponentially with growing measurements in scenarios with no other information present to distinguish the targets. Tensors, fundamentally, are just multi-dimensional arrays typically used to for data storage and transfer. By performing tensor decomposition, or high-dimensional principal component analysis (PCA), on a tensor it is reduced into two components — a low-dimensional summary known as the core, and a set of transformation matrices known as projection matrices. Dynamic Tensor Analysis (DTA) adapts this idea for a stream of data increasing in time. This research presents two methods based on tensor decomposition — Dynamic JPDA (DJPDA) and Windowed JPDA (WJPDA). The former employs DTA to reduce the dimensionality of the incoming set of measurements (scan) prior to association and reconstructs the associated set of measurements to obtain a more accurate track. Finally, this research also proposes an alternative approach to data association by treating it two separate machine learning problems — anomaly detection and measurement selection. The effectiveness of these algorithms are demonstrated with the help of 3 numerical problems — benchmark pedestrian tracking, ground robot tracking, and space debris tracking. The ground robot tracking involves the measurement and estimation of up to 5 objects (including clutter and random noise) in the field of view. This project is utilized as a test bed to check the computational efficiency of the proposed algorithm. Space debris tracking, a subset of the larger set of problems known as Space Situational Awareness (SSA), involves tracking multiple objects situated close to each other in the low Earth orbit (LEO). A critical aspect of SSA is the need to complete all necessary computations within a limited time due to the nature of observatory locations. In this simulation, it is assumed that a total of 6 observatories cover the orbital space which places a hard time bound for data association and estimation in each local station.
Mrinal Kumar (Advisor)
Levent Guvenc (Committee Member)
Ran Dai (Committee Member)
173 p.
Aerospace Engineering; Mechanical Engineering
tensor decomposition, data association, target tracking

Recommended Citations

Citations

  • Krishnaswamy, S. (2019). On Computationally Efficient Frameworks For Data Association In Multi-Target Tracking [Doctoral dissertation, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1574672274983947

    APA Style (7th edition)

  • Krishnaswamy, Sriram. On Computationally Efficient Frameworks For Data Association In Multi-Target Tracking. 2019. Ohio State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1574672274983947.

    MLA Style (8th edition)

  • Krishnaswamy, Sriram. "On Computationally Efficient Frameworks For Data Association In Multi-Target Tracking." Doctoral dissertation, Ohio State University, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=osu1574672274983947

    Chicago Manual of Style (17th edition)

osu1574672274983947
99
© 2019, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.