Date of Award

Spring 4-2021

Access Type

Thesis - Open Access

Degree Name

Master of Science in Electrical & Computer Engineering


Electrical Engineering and Computer Science

Committee Chair

M. Ilhan Akbas

First Committee Member

Patrick Currier

Second Committee Member

Eric Coyle


Multi-Object Tracking (MOT) is a field critical to Automated Vehicle (AV) perception systems. However, it is large, complex, spans research fields, and lacks resources for integration with real sensors and implementation on AVs. Factors such those make it difficult for new researchers and practitioners to enter the field.

This thesis presents two main contributions: 1) a comprehensive mapping for the field of Multi-Object Trackers (MOTs) with a specific focus towards Automated Vehicles (AVs) and 2) a real-world evaluation of an MOT developed and tuned using COTS (Commercial Off-The-Shelf) software toolsets. The first contribution aims to give a comprehensive overview of MOTs and various MOT subfields for AVs that have not been presented as wholistically in other papers. The second contribution aims to illustrate some of the benefits of using a COTS MOT toolset and some of the difficulties associated with using real-world data. This MOT performed accurate state estimation of a target vehicle through the tracking and fusion of data from a radar and vision sensor using a Central-Level Track Processing approach and a Global Nearest Neighbors assignment algorithm. It had an 0.44 m positional Root Mean Squared Error (RMSE) over a 40 m approach test.

It is the authors' hope that this work provides an overview of the MOT field that will help new researchers and practitioners enter the field. Additionally, the author hopes that the evaluation section illustrates some difficulties of using real-world data and provides a good pathway for developing and deploying MOTs from software toolsets to Automated Vehicles.