Retrieval-Based Learning for Efficient High-DoF Motion Planning

dc.contributor.advisorKavraki, Lydia Een_US
dc.contributor.advisorShrivastava, Anshumalien_US
dc.creatorChamzas, Constantinosen_US
dc.date.accessioned2023-08-09T18:48:54Zen_US
dc.date.available2023-08-09T18:48:54Zen_US
dc.date.created2023-05en_US
dc.date.issued2023-04-20en_US
dc.date.submittedMay 2023en_US
dc.date.updated2023-08-09T18:48:55Zen_US
dc.description.abstractSampling-based planners are effective in many real-world applications such as robotics manipulation, navigation, and even protein modeling. However, in the absence of any prior information, sampling-based planners are forced to explore uniformly or heuristically, which can lead to degraded performance. One way to improve performance is to use prior knowledge of environments to adapt the sampling strategy. In this thesis, we propose Pyre, a retrieval-based framework for learning sampling distributions for high-dimensional motion planning. Pyre decomposes the workspace into local primitives, memorizing local experiences by these primitives in the form of local samplers, and storing them in a database. Pyre synthesizes a global sampler by learning to retrieve local experiences relevant to the given situation. This method transfers knowledge between diverse environments that share local primitives and significantly speeds up planning. Pyre learns incrementally and significantly improves performance with only a handful of examples, achieving better generalization over diverse tasks and environments as compared to prior approaches. We demonstrate the effectiveness of Pyre on 2D and 3D environments as well as with a simulated and real Fetch Robot tasked with challenging pick-and-place manipulation problems. Additionally, in this thesis, we also contribute MotionBenchMaker, an open-source benchmarking tool for realistic robot manipulation problems. It is an extensible, easy-to-use tool that allows users to both generate datasets and benchmark them by comparing motion planning algorithms.en_US
dc.format.mimetypeapplication/pdfen_US
dc.identifier.citationChamzas, Constantinos. "Retrieval-Based Learning for Efficient High-DoF Motion Planning." (2023) Diss., Rice University. <a href="https://hdl.handle.net/1911/115150">https://hdl.handle.net/1911/115150</a>.en_US
dc.identifier.urihttps://hdl.handle.net/1911/115150en_US
dc.language.isoengen_US
dc.rightsCopyright is held by the author, unless otherwise indicated. Permission to reuse, publish, or reproduce the work beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder.en_US
dc.subjectRobotic Motion Planningen_US
dc.subjectMachine Learningen_US
dc.titleRetrieval-Based Learning for Efficient High-DoF Motion Planningen_US
dc.typeThesisen_US
dc.type.materialTexten_US
thesis.degree.departmentComputer Scienceen_US
thesis.degree.disciplineEngineeringen_US
thesis.degree.grantorRice Universityen_US
thesis.degree.levelDoctoralen_US
thesis.degree.nameDoctor of Philosophyen_US
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
CHAMZAS-DOCUMENT-2023.pdf
Size:
19.08 MB
Format:
Adobe Portable Document Format
License bundle
Now showing 1 - 2 of 2
No Thumbnail Available
Name:
PROQUEST_LICENSE.txt
Size:
5.85 KB
Format:
Plain Text
Description:
No Thumbnail Available
Name:
LICENSE.txt
Size:
2.61 KB
Format:
Plain Text
Description: