Tactile regrasp of objects with dynamic center-of-mass
Many household objects have a container-like shape, with contents that can move inside. When the contents are heavier than the container, their movement can cause the object’s center of mass (CoM) to shift. If a robot grasps the object far from the CoM, this can induce rotation and create a moving t...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/167487 |
| _version_ | 1826128609421557760 |
|---|---|
| author | Than, Duc Huy |
| author2 | Lin Zhiping |
| author_facet | Lin Zhiping Than, Duc Huy |
| author_sort | Than, Duc Huy |
| collection | NTU |
| description | Many household objects have a container-like shape, with contents that can move inside. When the contents are heavier than the container, their movement can cause the object’s center of mass (CoM) to shift. If a robot grasps the object far from the CoM, this can induce rotation and create a moving target that is difficult for a control policy to track. This project proposes a regrasp policy that utilizes the Gelsight tactile sensor to train a stability classifier and value-based DQN agent. The goal is to enable the robot to grasp objects with dynamic CoM with as few regrasps as possible. The proposed approach employs offline Reinforcement Learning (RL) to achieve sample efficiency during the data collection and training process. Various design choices and hyperparameters of the RL agents are explored and the best-performing agent exhibits high adaptability, due to its ability to adjust the step size in each regrasp attempt. |
| first_indexed | 2024-10-01T07:27:43Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/167487 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T07:27:43Z |
| publishDate | 2023 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1674872023-07-07T15:47:15Z Tactile regrasp of objects with dynamic center-of-mass Than, Duc Huy Lin Zhiping School of Electrical and Electronic Engineering Institute for Infocomm Research, A*STAR King's College London EZPLin@ntu.edu.sg Engineering::Electrical and electronic engineering::Control and instrumentation::Robotics Many household objects have a container-like shape, with contents that can move inside. When the contents are heavier than the container, their movement can cause the object’s center of mass (CoM) to shift. If a robot grasps the object far from the CoM, this can induce rotation and create a moving target that is difficult for a control policy to track. This project proposes a regrasp policy that utilizes the Gelsight tactile sensor to train a stability classifier and value-based DQN agent. The goal is to enable the robot to grasp objects with dynamic CoM with as few regrasps as possible. The proposed approach employs offline Reinforcement Learning (RL) to achieve sample efficiency during the data collection and training process. Various design choices and hyperparameters of the RL agents are explored and the best-performing agent exhibits high adaptability, due to its ability to adjust the step size in each regrasp attempt. Bachelor of Engineering (Electrical and Electronic Engineering) 2023-05-29T07:23:18Z 2023-05-29T07:23:18Z 2023 Final Year Project (FYP) Than, D. H. (2023). Tactile regrasp of objects with dynamic center-of-mass. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/167487 https://hdl.handle.net/10356/167487 en CY3001-222 application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Electrical and electronic engineering::Control and instrumentation::Robotics Than, Duc Huy Tactile regrasp of objects with dynamic center-of-mass |
| title | Tactile regrasp of objects with dynamic center-of-mass |
| title_full | Tactile regrasp of objects with dynamic center-of-mass |
| title_fullStr | Tactile regrasp of objects with dynamic center-of-mass |
| title_full_unstemmed | Tactile regrasp of objects with dynamic center-of-mass |
| title_short | Tactile regrasp of objects with dynamic center-of-mass |
| title_sort | tactile regrasp of objects with dynamic center of mass |
| topic | Engineering::Electrical and electronic engineering::Control and instrumentation::Robotics |
| url | https://hdl.handle.net/10356/167487 |
| work_keys_str_mv | AT thanduchuy tactileregraspofobjectswithdynamiccenterofmass |