Off-campus Michigan Tech users: To download campus access theses or dissertations, please use the following button to log in with your Michigan Tech ID and password: log in to proxy server

Non-Michigan Tech users: Please talk to your librarian about requesting this thesis or dissertation through interlibrary loan.

Date of Award


Document Type

Campus Access Dissertation

Degree Name

Doctor of Philosophy in Electrical Engineering (PhD)

Administrative Home Department

Department of Electrical and Computer Engineering

Advisor 1

Saeid Nooshabadi

Advisor 2

Glen Archer

Committee Member 1

Daniel Fuhrmann

Committee Member 2

Jeremy Bos


SUVs and pickup trucks have a significant share of the United States market. This fact has motivated automakers to equip their products with more safety and convenience features, such as the Advanced trailer backup assistance system (TBAS). TBAS helps to take the frustration out of the driver of the tow-vehicle during the backup maneuvers. To operate an autonomous reversing maneuver, the TBAS requires the trailer-tow vehicle combined kinematic model, where a key parameter is the articulation angle between the tow vehicle and the trailer.

In this dissertation, we aim to develop three models to detect the articulation angle between tow-vehicle and trailer using the rear-side camera and radar sensors. Models, each designed as an independent module for detecting the trailer articulation angle, are; computer vision-based hitch angle detection, radar-based hitch angle tracking, and deep learning-based RadarRegNet-based hitch angle estimation.

The computer vision-based module estimates the relative trailer angle using a deep learning object detection model to detect marker-lights on the trailer. The proposed computer vision model processes the image frames acquired from the rear-facing camera to detect and track the trailer and estimate its orientation to the tow vehicle.

The proposed radar-based hitch angle tracking, for the estimate of the hitch angle, processes reflections acquired from the mmWave radars situated at the rear side of the vehicle to detect the trailer and track its orientation in relation to the tow-vehicle. This technique is based on the tracking of individual points in the merged radars point-cloud. Each tracked point is considered as a hitch angle estimator. Using the current and past position information of a point, the model estimates the current hitch angle.

Finally, the proposed RadarRegNet-based hitch angle estimation model employs a deep learning image regression convolutional neural network. The network input is an occupancy grid map of the radar points in the merged point-clouds acquired by two radars, and its output is the hitch angle. Aiming for a reasonable inference time for the time-critical hitch angle estimation task, we introduced a modified Inception block. In the proposed modified Inception block to reduce the computational cost by up to 50%.