Honors Program
Date of this Version
5-2022
Document Type
Thesis
Citation
Derowitsch, L., R. Nordgren, K. Karkazis, S. Patel, D. Niyonshuti, S. Schneider, C. Pirisingula. Drone Progress Tracking Application. Undergraduate
Honors Thesis. University of Nebraska-Lincoln. 2022.
Abstract
Kiewit is one of the largest construction companies in the world and has long pushed the standards for innovation within the construction industry. Kiewit operates on an initiative to be a data-driven organization. Construction projects track progress data for a variety of reasons, none more important than meeting contractual obligations. Keeping clients in tune with the status of a given project not only gives them peace of mind but also correlates directly with revenue. For solar projects, quantity claiming is done by walking up and down rows of solar panels, posts, and torque tubes across the project site and manually entering the status data. The project goal was to create a secondary automated version of the manual quantity claiming process for construction progress tracking.
The Kiewit Drone Progress Tracking application serves as a digital management system that supports tracking the progress of solar panel construction sites in a more streamlined way. It hosts a machine learning model that was built in-house. It predicts the progress of the construction site based on drone-captured geo-location tagged images called GeoTIFFs. The application hosts these GeoTIFFs and allows for the creation and display of labels that ultimately help train and run the machine learning model. The development of this application will increase the training data used to impact the accuracy of the AI model, as well as improve accessibility and efficiency for the current solution and provide the platform for expansion to other construction project types.
Comments
Copyright Laura Derowitsch, Rachel Nordgren, Kally Karkazis, Savan Patel, Damien Niyonshuti, Sam Schneider & Chaitra Pirisingula 2022