{"id":{"repo_id":"msu","oai_identifier":"oai:d.lib.msu.edu:etd_53066"},"canonical_url":"https://search.ndltd.org/etd/msu/oai:d.lib.msu.edu:etd_53066","repository":{"repo_id":"msu","name":"Michigan State University","base_url":"https://d.lib.msu.edu/oai"},"display":{"title":"Automated LiDAR-Camera Calibration for Autonomous Driving under Domain Shift due to Changing Weather Conditions","abstract":"Thesis (M.S.)--Michigan State University. Electrical and Computer Engineering - Master of Science, 2026","abstract_html":"Thesis (M.S.)--Michigan State University. Electrical and Computer Engineering - Master of Science, 2026","abstract_has_math":false,"creators":["Guo, Xinwei"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Radha, Hayder","Morris, Daniel","Traganitis, Panagiotis"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-11-18T18:59:37Z","subjects":["Engineering"],"languages":["eng"],"rights":["In Copyright"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:53066","local:Guo_grad.msu_0128N_22076"],"render_values":[{"text":"etd:53066","href":null,"code":true},{"text":"local:Guo_grad.msu_0128N_22076","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/doi:10.25335/3jqz-0377","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/doi:10.25335/3jqz-0377","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://d.lib.msu.edu/oai?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Ad.lib.msu.edu%3Aetd_53066","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Radha, Hayder","Morris, Daniel","Traganitis, Panagiotis"]},{"key":"dc:creator","label":"Author","values":["Guo, Xinwei"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2026"]},{"key":"dc:relation","label":"Dc Relation","values":["Electronic Theses & Dissertations"]},{"key":"dc:type","label":"Dc Type","values":["text","Electronic dissertations"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Engineering"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["In Copyright"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:53066","local:Guo_grad.msu_0128N_22076","https://doi.org/doi:10.25335/3jqz-0377"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Thesis (M.S.)--Michigan State University. Electrical and Computer Engineering - Master of Science, 2026","With the rapid development of autonomous driving technology, multi-sensor fusion techniques have become essential for achieving high-precision perception and understanding of vehicles' surroundings. In this process, accurate sensor extrinsic calibration has become ain critical prerequisite, and it is of great importance to develop robust and accurate automated LiDAR–camera calibration methods in diverse and challenging environments. This thesis first presents a systematic review of LiDAR–camera extrinsic calibration approaches, additionally, a unified evaluation framework is applied by simulating calibration perturbations on the KITTI dataset and comparing the performance of different calibration methods using a multi-modal 3D object detection network. Experimental results indicate that recent deep learning-based methods, in general, and the Lidar-Camera Calibration Network (LCCNet) framework, in particular, achieve superior calibration accuracy, but with limited generalization capability across different data distributions. To address this issue, we propose a domain adaptation-enhanced LiDAR–camera calibration network, termed Domain Adaptive Lidar-Camera Calibration Network (DA-LCCNet), built upon the existing LCCNet framework. The proposed method preserves the original feature extraction and matching architecture of LCCNet while incorporating a Domain Adaptation Network (DAN) to learn domain-invariant multimodal feature representations, thereby improving calibration stability under changing weather conditions. We explore different DAN architectures including progressive Multi-Layer Perceptron (MLP), Variational Information Bottleneck (VIB) based, and hybrid MLP-VIB networks. Cross-weather experiments conducted on the Michigan State University-Four Seasons (MSU-4S) dataset demonstrate that our proposed domain adaptive framework, DA-LCCNet, significantly reduces calibration errors under sunny and rainy conditions and improves the robustness and generalization capability of multi-modal fusion systems. Overall, this research advances targetless LiDAR–camera calibration through both systematic evaluation and methodological development, providing theoretical and practical solutions for reliable multi-sensor fusion perception in complex environments.Keywords: Autonomous Driving, LiDAR–camera Calibration, Multimodal Sensor Fusion, Object Detection, Deep Learning, Generalization, Domain Adaptation","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf","1 online resource (43 p.) : ill."]},{"key":"dc:title","label":"Title","values":["Automated LiDAR-Camera Calibration for Autonomous Driving under Domain Shift due to Changing Weather Conditions"]}]}],"canonical_facts":{"dc:contributor":["Radha, Hayder","Morris, Daniel","Traganitis, Panagiotis"],"dc:creator":["Guo, Xinwei"],"dc:date":["2026"],"dc:description":["Thesis (M.S.)--Michigan State University. Electrical and Computer Engineering - Master of Science, 2026","With the rapid development of autonomous driving technology, multi-sensor fusion techniques have become essential for achieving high-precision perception and understanding of vehicles' surroundings. In this process, accurate sensor extrinsic calibration has become ain critical prerequisite, and it is of great importance to develop robust and accurate automated LiDAR–camera calibration methods in diverse and challenging environments. This thesis first presents a systematic review of LiDAR–camera extrinsic calibration approaches, additionally, a unified evaluation framework is applied by simulating calibration perturbations on the KITTI dataset and comparing the performance of different calibration methods using a multi-modal 3D object detection network. Experimental results indicate that recent deep learning-based methods, in general, and the Lidar-Camera Calibration Network (LCCNet) framework, in particular, achieve superior calibration accuracy, but with limited generalization capability across different data distributions. To address this issue, we propose a domain adaptation-enhanced LiDAR–camera calibration network, termed Domain Adaptive Lidar-Camera Calibration Network (DA-LCCNet), built upon the existing LCCNet framework. The proposed method preserves the original feature extraction and matching architecture of LCCNet while incorporating a Domain Adaptation Network (DAN) to learn domain-invariant multimodal feature representations, thereby improving calibration stability under changing weather conditions. We explore different DAN architectures including progressive Multi-Layer Perceptron (MLP), Variational Information Bottleneck (VIB) based, and hybrid MLP-VIB networks. Cross-weather experiments conducted on the Michigan State University-Four Seasons (MSU-4S) dataset demonstrate that our proposed domain adaptive framework, DA-LCCNet, significantly reduces calibration errors under sunny and rainy conditions and improves the robustness and generalization capability of multi-modal fusion systems. Overall, this research advances targetless LiDAR–camera calibration through both systematic evaluation and methodological development, providing theoretical and practical solutions for reliable multi-sensor fusion perception in complex environments.Keywords: Autonomous Driving, LiDAR–camera Calibration, Multimodal Sensor Fusion, Object Detection, Deep Learning, Generalization, Domain Adaptation","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."],"dc:format":["application/pdf","1 online resource (43 p.) : ill."],"dc:identifier":["etd:53066","local:Guo_grad.msu_0128N_22076","https://doi.org/doi:10.25335/3jqz-0377"],"dc:language":["eng"],"dc:relation":["Electronic Theses & Dissertations"],"dc:rights":["In Copyright"],"dc:subject":["Engineering"],"dc:title":["Automated LiDAR-Camera Calibration for Autonomous Driving under Domain Shift due to Changing Weather Conditions"],"dc:type":["text","Electronic dissertations"]},"updated_at":"2026-11-18T18:59:37Z"}