{"id":{"repo_id":"mit","oai_identifier":"oai:dspace.mit.edu:1721.1/164567"},"canonical_url":"https://search.dev.ndltd.org/etd/mit/oai:dspace.mit.edu:1721.1/164567","repository":{"repo_id":"mit","name":"MIT","base_url":"https://dspace.mit.edu/oai/request"},"display":{"title":"Generalizable Robot Manipulation through Unified Perception, Policy Learning, and Planning","abstract":"Advancing robotic manipulation to achieve generalization across diverse goals, environments, and embodiments is a critical challenge in robotics research. While the availability of data and large-scale training has brought exciting progress in robotics manipulation, current methods often struggle with generalizing to unseen, unstructured environments and solving long-horizon tasks. In this thesis, I will present my work in robot learning and planning that enables multi-step manipulation in partially observable environments, towards general-purpose embodied agents. Specifically, I will talk about my work in 1) constructing a modular framework that estimates affordances with learned perception models with task-and-motion-planning (TAMP) for object rearrangement in unstructured scenes, 2) learning generative diffusion models of robot skills, which can be composed to solve unseen combination of environmental constraints through infeference-time optimization, 3) leveraging large vision-language models (VLMs) in building task-oriented visual abstractions, allowing skills to generalize across different environments with only 5 to 10 demonstrations. Together, these approaches contribute to the generality and scalability of embodied agents towards solving real-world manipulation in unstructured environments.","abstract_html":"Advancing robotic manipulation to achieve generalization across diverse goals, environments, and embodiments is a critical challenge in robotics research. While the availability of data and large-scale training has brought exciting progress in robotics manipulation, current methods often struggle with generalizing to unseen, unstructured environments and solving long-horizon tasks. In this thesis, I will present my work in robot learning and planning that enables multi-step manipulation in partially observable environments, towards general-purpose embodied agents. Specifically, I will talk about my work in 1) constructing a modular framework that estimates affordances with learned perception models with task-and-motion-planning (TAMP) for object rearrangement in unstructured scenes, 2) learning generative diffusion models of robot skills, which can be composed to solve unseen combination of environmental constraints through infeference-time optimization, 3) leveraging large vision-language models (VLMs) in building task-oriented visual abstractions, allowing skills to generalize across different environments with only 5 to 10 demonstrations. Together, these approaches contribute to the generality and scalability of embodied agents towards solving real-world manipulation in unstructured environments.","abstract_has_math":false,"creators":["Fang, Xiaolin"],"institution":"Massachusetts Institute of Technology","degree_name":"Doctoral","degree_level":null,"degree_discipline":null,"degree_department":"Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science","school":null,"contributors":[],"advisors":["Kaelbling, Leslie Pack","Lozano-Pérez, Tomás"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-09","date_published":"2025-09","updated_at":"2026-07-22T22:21:05Z","subjects":[],"languages":[],"rights":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"],"rights_urls":["https://rightsstatements.org/page/InC-EDU/1.0/"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1721.1/164567","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Kaelbling, Leslie Pack","Lozano-Pérez, Tomás"]},{"key":"dc:contributor.department","label":"Department","values":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"]},{"key":"dc:creator","label":"Author","values":["Fang, Xiaolin"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-01-20T19:45:45Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2026-01-20T19:45:45Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-09"]},{"key":"dc:publisher","label":"Institution","values":["Massachusetts Institute of Technology"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctoral","Doctor of Philosophy"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://rightsstatements.org/page/InC-EDU/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1721.1/164567"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Advancing robotic manipulation to achieve generalization across diverse goals, environments, and embodiments is a critical challenge in robotics research. While the availability of data and large-scale training has brought exciting progress in robotics manipulation, current methods often struggle with generalizing to unseen, unstructured environments and solving long-horizon tasks. In this thesis, I will present my work in robot learning and planning that enables multi-step manipulation in partially observable environments, towards general-purpose embodied agents. Specifically, I will talk about my work in 1) constructing a modular framework that estimates affordances with learned perception models with task-and-motion-planning (TAMP) for object rearrangement in unstructured scenes, 2) learning generative diffusion models of robot skills, which can be composed to solve unseen combination of environmental constraints through infeference-time optimization, 3) leveraging large vision-language models (VLMs) in building task-oriented visual abstractions, allowing skills to generalize across different environments with only 5 to 10 demonstrations. Together, these approaches contribute to the generality and scalability of embodied agents towards solving real-world manipulation in unstructured environments."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Ph.D."]},{"key":"dc:title","label":"Title","values":["Generalizable Robot Manipulation through Unified Perception, Policy Learning, and Planning"]}]}],"canonical_facts":{"dc:contributor.advisor":["Kaelbling, Leslie Pack","Lozano-Pérez, Tomás"],"dc:contributor.department":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"],"dc:creator":["Fang, Xiaolin"],"dc:date.accessioned":["2026-01-20T19:45:45Z"],"dc:date.available":["2026-01-20T19:45:45Z"],"dc:date.issued":["2025-09"],"dc:description.abstract":["Advancing robotic manipulation to achieve generalization across diverse goals, environments, and embodiments is a critical challenge in robotics research. While the availability of data and large-scale training has brought exciting progress in robotics manipulation, current methods often struggle with generalizing to unseen, unstructured environments and solving long-horizon tasks. In this thesis, I will present my work in robot learning and planning that enables multi-step manipulation in partially observable environments, towards general-purpose embodied agents. Specifically, I will talk about my work in 1) constructing a modular framework that estimates affordances with learned perception models with task-and-motion-planning (TAMP) for object rearrangement in unstructured scenes, 2) learning generative diffusion models of robot skills, which can be composed to solve unseen combination of environmental constraints through infeference-time optimization, 3) leveraging large vision-language models (VLMs) in building task-oriented visual abstractions, allowing skills to generalize across different environments with only 5 to 10 demonstrations. Together, these approaches contribute to the generality and scalability of embodied agents towards solving real-world manipulation in unstructured environments."],"dc:description.degree":["Ph.D."],"dc:identifier.uri":["https://hdl.handle.net/1721.1/164567"],"dc:publisher":["Massachusetts Institute of Technology"],"dc:rights":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"],"dc:rights.uri":["https://rightsstatements.org/page/InC-EDU/1.0/"],"dc:title":["Generalizable Robot Manipulation through Unified Perception, Policy Learning, and Planning"],"dc:type":["Thesis"],"thesis:degree_name":["Doctoral","Doctor of Philosophy"]},"updated_at":"2026-07-22T22:21:05Z"}