{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/108536"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/108536","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Hierarchical reinforcement learning for adaptive and autonomous decision-making in robotics","abstract":"In recent years, Reinforcement Learning has been able to solve extremely complex games in simulation, but with limited success in deployment to real-world scenarios. The goal of this work is create an ecosystem in which Reinforcement Learning algorithms can be deployed onto real robots in complex games. The ecosystem begins with the creation of a development pipeline which can be used to progressively train Reinforcement Learning Algorithms in increasingly realistic scenarios, culminating with the deployment of these algorithm onto a real system. The pipeline is paired with the novel Reinforcement Learning algorithms that are better able to adapt to new scenarios than traditional methods for autonomy and robotic planning.We implement two techniques to enable this adaptation. First, we implement a hierarchical Reinforcement Learning architecture that uses differentiated sub-policies governed by a hierarchical controller to enable fast adaptation. Second we introduce a confidence-based training process for the hierarchical controller which improves training stability and convergence times. These algorithmic contributions were evaluated using our development pipeline.","abstract_html":"In recent years, Reinforcement Learning has been able to solve extremely complex games in simulation, but with limited success in deployment to real-world scenarios. The goal of this work is create an ecosystem in which Reinforcement Learning algorithms can be deployed onto real robots in complex games. The ecosystem begins with the creation of a development pipeline which can be used to progressively train Reinforcement Learning Algorithms in increasingly realistic scenarios, culminating with the deployment of these algorithm onto a real system. The pipeline is paired with the novel Reinforcement Learning algorithms that are better able to adapt to new scenarios than traditional methods for autonomy and robotic planning.We implement two techniques to enable this adaptation. First, we implement a hierarchical Reinforcement Learning architecture that uses differentiated sub-policies governed by a hierarchical controller to enable fast adaptation. Second we introduce a confidence-based training process for the hierarchical controller which improves training stability and convergence times. These algorithmic contributions were evaluated using our development pipeline.","abstract_has_math":false,"creators":["Van Stralen, Neale A"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Aerospace Engineering","degree_department":null,"school":null,"contributors":["Tran, Huy","Chowdhary, Girish"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2020,"date_issued":"2020-10-07T21:00:09Z","date_published":"2020-10-07T21:00:09Z","updated_at":"2026-07-22T22:24:48Z","subjects":["Reinforcement Learning","Adaptation"],"languages":["en"],"rights":["Copyright 2020 Neale Van Stralen"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/108536","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Tran, Huy","Chowdhary, Girish"]},{"key":"dc:creator","label":"Author","values":["Van Stralen, Neale A"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2020-10-07T21:00:09Z","2020-07-23","2020-08"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Aerospace Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Reinforcement Learning","Adaptation"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2020 Neale Van Stralen"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/108536"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["In recent years, Reinforcement Learning has been able to solve extremely complex games in simulation, but with limited success in deployment to real-world scenarios. The goal of this work is create an ecosystem in which Reinforcement Learning algorithms can be deployed onto real robots in complex games. The ecosystem begins with the creation of a development pipeline which can be used to progressively train Reinforcement Learning Algorithms in increasingly realistic scenarios, culminating with the deployment of these algorithm onto a real system. The pipeline is paired with the novel Reinforcement Learning algorithms that are better able to adapt to new scenarios than traditional methods for autonomy and robotic planning.We implement two techniques to enable this adaptation. First, we implement a hierarchical Reinforcement Learning architecture that uses differentiated sub-policies governed by a hierarchical controller to enable fast adaptation. Second we introduce a confidence-based training process for the hierarchical controller which improves training stability and convergence times. These algorithmic contributions were evaluated using our development pipeline.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2020-10-02 without embargo terms","The student, Neale Van Stralen, accepted the attached license on 2020-07-22 at 16:34.","The student, Neale Van Stralen, submitted this Thesis for approval on 2020-07-22 at 16:43.","This Thesis was approved for publication on 2020-07-23 at 13:07.","DSpace SAF Submission Ingestion Package generated from Vireo submission #15728 on 2020-10-02 at 15:15:19","Made available in DSpace on 2020-10-07T21:00:09Z (GMT). No. of bitstreams: 2 VANSTRALEN-THESIS-2020.pdf: 4065322 bytes, checksum: 727aab740ec4e2c5a2a9e71dac0b089b (MD5) LICENSE.txt: 4214 bytes, checksum: 637186641e2b77d7520a22865039306c (MD5) Previous issue date: 2020-07-23"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Hierarchical reinforcement learning for adaptive and autonomous decision-making in robotics"]}]}],"canonical_facts":{"dc:contributor":["Tran, Huy","Chowdhary, Girish"],"dc:creator":["Van Stralen, Neale A"],"dc:date":["2020-10-07T21:00:09Z","2020-07-23","2020-08"],"dc:description":["In recent years, Reinforcement Learning has been able to solve extremely complex games in simulation, but with limited success in deployment to real-world scenarios. The goal of this work is create an ecosystem in which Reinforcement Learning algorithms can be deployed onto real robots in complex games. The ecosystem begins with the creation of a development pipeline which can be used to progressively train Reinforcement Learning Algorithms in increasingly realistic scenarios, culminating with the deployment of these algorithm onto a real system. The pipeline is paired with the novel Reinforcement Learning algorithms that are better able to adapt to new scenarios than traditional methods for autonomy and robotic planning.We implement two techniques to enable this adaptation. First, we implement a hierarchical Reinforcement Learning architecture that uses differentiated sub-policies governed by a hierarchical controller to enable fast adaptation. Second we introduce a confidence-based training process for the hierarchical controller which improves training stability and convergence times. These algorithmic contributions were evaluated using our development pipeline.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2020-10-02 without embargo terms","The student, Neale Van Stralen, accepted the attached license on 2020-07-22 at 16:34.","The student, Neale Van Stralen, submitted this Thesis for approval on 2020-07-22 at 16:43.","This Thesis was approved for publication on 2020-07-23 at 13:07.","DSpace SAF Submission Ingestion Package generated from Vireo submission #15728 on 2020-10-02 at 15:15:19","Made available in DSpace on 2020-10-07T21:00:09Z (GMT). No. of bitstreams: 2 VANSTRALEN-THESIS-2020.pdf: 4065322 bytes, checksum: 727aab740ec4e2c5a2a9e71dac0b089b (MD5) LICENSE.txt: 4214 bytes, checksum: 637186641e2b77d7520a22865039306c (MD5) Previous issue date: 2020-07-23"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/108536"],"dc:language":["en"],"dc:rights":["Copyright 2020 Neale Van Stralen"],"dc:subject":["Reinforcement Learning","Adaptation"],"dc:title":["Hierarchical reinforcement learning for adaptive and autonomous decision-making in robotics"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Aerospace Engineering"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:24:48Z"}