2026-09-20T00:01:53Z https://www.ideals.illinois.edu/oai-pmh

oai:www.ideals.illinois.edu:2142/108550 2023-07-11 col_2142_5131 col_2142_14800 com_2142_5130 com_2142_14799 com_2142_234

2020-07-24 Chowdhary, Girish Havens, Aaron 2020-10-07T21:00:13Z 2020-10-07T21:00:13Z 2020-08 Methods like deep reinforcement learning (DRL) have gained increasing attention when solving very general continuous control tasks in a model-free end-to-end fashion. However, there has been great difficulty in applying these algorithms to real-world systems due to poor sample efficiency and inability to handle state and control constraints. We introduce and demonstrate a general paradigm that combines model-learning and online planning for control which can also handle a wide range of problems using traditional and non-traditional sensor information. Rather than using popular RL methods, learning a model from data and performing online planning in the form of model predictive control (MPC) can be much more data-efficient and practical for deploying on real robotics systems. In addition to a generally applicable sample-based planning strategy, another specific formulation of model learning is investigated that allows for a linear structure to be exploited for efficient control. The algorithms are validated in both simulation and on real robotic platforms, namely an agriculture berry-picking robot using a soft-continuum arm. The model-based method is not only able to solve a challenging soft-body control task, but also can be deployed in a field setting where model-free RL is bottle-necked by data-efficiency. Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2020-10-02 without embargo terms The student, Aaron Havens, accepted the attached license on 2020-07-24 at 15:11. The student, Aaron Havens, submitted this Thesis for approval on 2020-07-24 at 15:49. This Thesis was approved for publication on 2020-07-24 at 16:11. DSpace SAF Submission Ingestion Package generated from Vireo submission #15754 on 2020-10-02 at 15:15:32 Made available in DSpace on 2020-10-07T21:00:13Z (GMT). No. of bitstreams: 3 HAVENS-THESIS-2020.pdf: 7780017 bytes, checksum: 9fd22664d9aad42c295397d41db703ca (MD5) Havens_MS_Thesis.zip: 6479413 bytes, checksum: d91bed0dbca7582f628c1bed88f051b4 (MD5) LICENSE.txt: 4209 bytes, checksum: b1efb8edb2ee5f861708e517895ec56a (MD5) Previous issue date: 2020-07-24 application/pdf http://hdl.handle.net/2142/108550 en Copyright 2020 Aaron Havens model predictive control reinforcement learning soft robotics Model-based approaches for learning control from multi-modal data text Thesis Aerospace Engineering Aerospace Engineering University of Illinois at Urbana-Champaign Thesis M.S.