{"id":{"repo_id":"calpoly","oai_identifier":"oai:digitalcommons.calpoly.edu:theses-4860"},"canonical_url":"https://search.ndltd.org/etd/calpoly/oai:digitalcommons.calpoly.edu:theses-4860","repository":{"repo_id":"calpoly","name":"Cal Poly","base_url":"https://digitalcommons.calpoly.edu/do/oai/"},"display":{"title":"Hallucination Techniques for Self-Supervised Synthetic Datasets for Mobile Robots","abstract":"Classical techniques in autonomous navigation struggle in tightly constrained spaces. Machine learning has been shown to perform better in these difficult environments but most techniques require large amounts of navigation experience for training. Using a new machine learning paradigm learning from hallucination (LfH), training data can be collected in a safe environment and not require supervision. Data is collected in real time while an agent performs a random walk in free space, supervision is not required as there are no obstacles for the robot to run into. After a random walk a post processing pipeline will hallucinate a safety corridor around the path the robot performed during its random walk. This will trick a learning policy into believing the labels are a result of avoiding an obstacle. Our experiments show that variations in the creation of the safety corridor can improve model performance. Our approach enables a robot to navigate previously unseen environments with real obstacles despite only being trained in a simulated environment, alleviating the difficulties of real world demonstrations.","abstract_html":"Classical techniques in autonomous navigation struggle in tightly constrained spaces. Machine learning has been shown to perform better in these difficult environments but most techniques require large amounts of navigation experience for training. Using a new machine learning paradigm learning from hallucination (LfH), training data can be collected in a safe environment and not require supervision. Data is collected in real time while an agent performs a random walk in free space, supervision is not required as there are no obstacles for the robot to run into. After a random walk a post processing pipeline will hallucinate a safety corridor around the path the robot performed during its random walk. This will trick a learning policy into believing the labels are a result of avoiding an obstacle. Our experiments show that variations in the creation of the safety corridor can improve model performance. Our approach enables a robot to navigate previously unseen environments with real obstacles despite only being trained in a simulated environment, alleviating the difficulties of real world demonstrations.","abstract_has_math":false,"creators":["Colburn, Wyatt D"],"institution":null,"degree_name":"MS in Engineering","degree_level":null,"degree_discipline":"Electrical Engineering","degree_department":null,"school":null,"contributors":["Carlos Diaz Alvarenga","Electrical Engineering","College of Engineering"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T07:12:41Z","subjects":["Autonomous navigation","imitation learning","machine learning for robot control","neural networks in local planning","Controls and Control Theory","Robotics"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.calpoly.edu/theses/3216","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://digitalcommons.calpoly.edu/theses/3216","record_label":"Repository record","record_kind":"record","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://digitalcommons.calpoly.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.calpoly.edu%3Atheses-4860","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Carlos Diaz Alvarenga","Electrical Engineering","College of Engineering"]},{"key":"dc:creator","label":"Author","values":["Colburn, Wyatt D"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2025-12-10T08:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2025-12-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical Engineering"]},{"key":"thesis:degree_name","label":"Degree Name","values":["MS in Engineering"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Autonomous navigation","imitation learning","machine learning for robot control","neural networks in local planning","Controls and Control Theory","Robotics"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.calpoly.edu/theses/3216"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Classical techniques in autonomous navigation struggle in tightly constrained spaces. Machine learning has been shown to perform better in these difficult environments but most techniques require large amounts of navigation experience for training. Using a new machine learning paradigm learning from hallucination (LfH), training data can be collected in a safe environment and not require supervision. Data is collected in real time while an agent performs a random walk in free space, supervision is not required as there are no obstacles for the robot to run into. After a random walk a post processing pipeline will hallucinate a safety corridor around the path the robot performed during its random walk. This will trick a learning policy into believing the labels are a result of avoiding an obstacle. Our experiments show that variations in the creation of the safety corridor can improve model performance. Our approach enables a robot to navigate previously unseen environments with real obstacles despite only being trained in a simulated environment, alleviating the difficulties of real world demonstrations."]},{"key":"dc:title","label":"Title","values":["Hallucination Techniques for Self-Supervised Synthetic Datasets for Mobile Robots"]}]}],"canonical_facts":{"dc:contributor":["Carlos Diaz Alvarenga","Electrical Engineering","College of Engineering"],"dc:creator":["Colburn, Wyatt D"],"dc:date.available":["2025-12-10T08:00:00Z"],"dc:date.created":["2025-12-01T08:00:00Z"],"dc:description.abstract":["Classical techniques in autonomous navigation struggle in tightly constrained spaces. Machine learning has been shown to perform better in these difficult environments but most techniques require large amounts of navigation experience for training. Using a new machine learning paradigm learning from hallucination (LfH), training data can be collected in a safe environment and not require supervision. Data is collected in real time while an agent performs a random walk in free space, supervision is not required as there are no obstacles for the robot to run into. After a random walk a post processing pipeline will hallucinate a safety corridor around the path the robot performed during its random walk. This will trick a learning policy into believing the labels are a result of avoiding an obstacle. Our experiments show that variations in the creation of the safety corridor can improve model performance. Our approach enables a robot to navigate previously unseen environments with real obstacles despite only being trained in a simulated environment, alleviating the difficulties of real world demonstrations."],"dc:identifier":["https://digitalcommons.calpoly.edu/theses/3216"],"dc:subject":["Autonomous navigation","imitation learning","machine learning for robot control","neural networks in local planning","Controls and Control Theory","Robotics"],"dc:title":["Hallucination Techniques for Self-Supervised Synthetic Datasets for Mobile Robots"],"thesis:degree_discipline":["Electrical Engineering"],"thesis:degree_name":["MS in Engineering"]},"updated_at":"2026-11-19T07:12:41Z"}