{"id":{"repo_id":"msu","oai_identifier":"oai:d.lib.msu.edu:etd_3408"},"canonical_url":"https://search.ndltd.org/etd/msu/oai:d.lib.msu.edu:etd_3408","repository":{"repo_id":"msu","name":"Michigan State University","base_url":"https://d.lib.msu.edu/oai"},"display":{"title":"How multi-frame model fitting and differential measurements can improve lidar-based vehicle tracking accuracy","abstract":"ABSTRACTHOW MULTI-FRAME MODEL FITTING AND DIFFERENTIAL MEASUREMENTSCAN IMPROVE LIDAR-BASED VEHICLE TRACKING ACCURACYBySteven J. Chao Over the past few decades, much work has been done in the field of laser based (Lidar) vehicle tracking. The most common approach is to fit a simple rectangular model to a point cloud of vehicle data, and then process the measurements using an Extended Kalman Filter. In this study we explore the use of techniques which could improve tracking performance in situations with sparse data, system noise or clutter, difficult vehicle orientations, or poorly modeled vehicle shapes. We specifically consider: Multi-Frame Measurements and Differential Measurements. Multiple-Frame Model Fitting can improve tracking accuracy through batch processing of multiple frames rather than the usual single-frame processing. Differential Measurement tracking avoids estimating a vehicle's current pose and instead estimates change in pose between subsequent frames by comparing point clouds, and has the advantage that it does not require a prior vehicle shape model. Since our goal is to focus specifically on these types of vehicle measurement, we ignore other important tasks such as background removal, object detection and classification, vehicle occlusion, and real-time speed considerations. We ultimately show that each of these measurement techniques have different, complementary strengths, which can be combined to improve vehicle tracking performance in difficult situations.","abstract_html":"ABSTRACTHOW MULTI-FRAME MODEL FITTING AND DIFFERENTIAL MEASUREMENTSCAN IMPROVE LIDAR-BASED VEHICLE TRACKING ACCURACYBySteven J. Chao Over the past few decades, much work has been done in the field of laser based (Lidar) vehicle tracking. The most common approach is to fit a simple rectangular model to a point cloud of vehicle data, and then process the measurements using an Extended Kalman Filter. In this study we explore the use of techniques which could improve tracking performance in situations with sparse data, system noise or clutter, difficult vehicle orientations, or poorly modeled vehicle shapes. We specifically consider: Multi-Frame Measurements and Differential Measurements. Multiple-Frame Model Fitting can improve tracking accuracy through batch processing of multiple frames rather than the usual single-frame processing. Differential Measurement tracking avoids estimating a vehicle&#x27;s current pose and instead estimates change in pose between subsequent frames by comparing point clouds, and has the advantage that it does not require a prior vehicle shape model. Since our goal is to focus specifically on these types of vehicle measurement, we ignore other important tasks such as background removal, object detection and classification, vehicle occlusion, and real-time speed considerations. We ultimately show that each of these measurement techniques have different, complementary strengths, which can be combined to improve vehicle tracking performance in difficult situations.","abstract_has_math":false,"creators":["Chao, Steven J."],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Morris, Daniel","Khalil, Hassan","Tan, Xiaobo"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2015,"date_issued":"2015","date_published":"2015","updated_at":"2026-11-18T18:06:36Z","subjects":["Automated vehicles","Optical radar","Tracking (Engineering)"],"languages":["English"],"rights":["In Copyright"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:3408","isbn:9781339331539","isbn:1339331535","oclc:945891042","umi:1605557","local:Chao_grad.msu_0128N_14056"],"render_values":[{"text":"etd:3408","href":null,"code":true},{"text":"isbn:9781339331539","href":null,"code":true},{"text":"isbn:1339331535","href":null,"code":true},{"text":"oclc:945891042","href":null,"code":true},{"text":"umi:1605557","href":null,"code":true},{"text":"local:Chao_grad.msu_0128N_14056","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/doi:10.25335/0mz9-3c18","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/doi:10.25335/0mz9-3c18","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_3408","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Morris, Daniel","Khalil, Hassan","Tan, Xiaobo"]},{"key":"dc:creator","label":"Author","values":["Chao, Steven J."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2015"]},{"key":"dc:relation","label":"Dc Relation","values":["Electronic Theses & Dissertations"]},{"key":"dc:type","label":"Dc Type","values":["Text","Theses"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Automated vehicles","Optical radar","Tracking (Engineering)"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["English"]},{"key":"dc:rights","label":"Dc Rights","values":["In Copyright"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:3408","isbn:9781339331539","isbn:1339331535","oclc:945891042","umi:1605557","local:Chao_grad.msu_0128N_14056","https://doi.org/doi:10.25335/0mz9-3c18"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["ABSTRACTHOW MULTI-FRAME MODEL FITTING AND DIFFERENTIAL MEASUREMENTSCAN IMPROVE LIDAR-BASED VEHICLE TRACKING ACCURACYBySteven J. Chao Over the past few decades, much work has been done in the field of laser based (Lidar) vehicle tracking. The most common approach is to fit a simple rectangular model to a point cloud of vehicle data, and then process the measurements using an Extended Kalman Filter. In this study we explore the use of techniques which could improve tracking performance in situations with sparse data, system noise or clutter, difficult vehicle orientations, or poorly modeled vehicle shapes. We specifically consider: Multi-Frame Measurements and Differential Measurements. Multiple-Frame Model Fitting can improve tracking accuracy through batch processing of multiple frames rather than the usual single-frame processing. Differential Measurement tracking avoids estimating a vehicle's current pose and instead estimates change in pose between subsequent frames by comparing point clouds, and has the advantage that it does not require a prior vehicle shape model. Since our goal is to focus specifically on these types of vehicle measurement, we ignore other important tasks such as background removal, object detection and classification, vehicle occlusion, and real-time speed considerations. We ultimately show that each of these measurement techniques have different, complementary strengths, which can be combined to improve vehicle tracking performance in difficult situations.","Thesis (M.S.)--Michigan State University. Electrical Engineering, 2015","Includes bibliographical references (pages 60-61)"]},{"key":"dc:format","label":"Dc Format","values":["ix, 61 pages","application/pdf"]},{"key":"dc:title","label":"Title","values":["How multi-frame model fitting and differential measurements can improve lidar-based vehicle tracking accuracy"]}]}],"canonical_facts":{"dc:contributor":["Morris, Daniel","Khalil, Hassan","Tan, Xiaobo"],"dc:creator":["Chao, Steven J."],"dc:date":["2015"],"dc:description":["ABSTRACTHOW MULTI-FRAME MODEL FITTING AND DIFFERENTIAL MEASUREMENTSCAN IMPROVE LIDAR-BASED VEHICLE TRACKING ACCURACYBySteven J. Chao Over the past few decades, much work has been done in the field of laser based (Lidar) vehicle tracking. The most common approach is to fit a simple rectangular model to a point cloud of vehicle data, and then process the measurements using an Extended Kalman Filter. In this study we explore the use of techniques which could improve tracking performance in situations with sparse data, system noise or clutter, difficult vehicle orientations, or poorly modeled vehicle shapes. We specifically consider: Multi-Frame Measurements and Differential Measurements. Multiple-Frame Model Fitting can improve tracking accuracy through batch processing of multiple frames rather than the usual single-frame processing. Differential Measurement tracking avoids estimating a vehicle's current pose and instead estimates change in pose between subsequent frames by comparing point clouds, and has the advantage that it does not require a prior vehicle shape model. Since our goal is to focus specifically on these types of vehicle measurement, we ignore other important tasks such as background removal, object detection and classification, vehicle occlusion, and real-time speed considerations. We ultimately show that each of these measurement techniques have different, complementary strengths, which can be combined to improve vehicle tracking performance in difficult situations.","Thesis (M.S.)--Michigan State University. Electrical Engineering, 2015","Includes bibliographical references (pages 60-61)"],"dc:format":["ix, 61 pages","application/pdf"],"dc:identifier":["etd:3408","isbn:9781339331539","isbn:1339331535","oclc:945891042","umi:1605557","local:Chao_grad.msu_0128N_14056","https://doi.org/doi:10.25335/0mz9-3c18"],"dc:language":["English"],"dc:relation":["Electronic Theses & Dissertations"],"dc:rights":["In Copyright"],"dc:subject":["Automated vehicles","Optical radar","Tracking (Engineering)"],"dc:title":["How multi-frame model fitting and differential measurements can improve lidar-based vehicle tracking accuracy"],"dc:type":["Text","Theses"]},"updated_at":"2026-11-18T18:06:36Z"}