Robot Reinforcement Learning for Object Isolation
(2024)Department of Automatic Control
- Abstract
- This thesis employs deep reinforcement learning, a branch of machine learning, to carry out robotic tasks. The objective centers around teaching an agent controlling a seven-axis robot arm with a gripper tool, to complete an object isolation task. For this task, a robot manipulates a cluttered environment in such a way that a predetermined target object becomes isolated. Sub-tasks were developed to explore simpler robot tasks to evolve and combine into more complex tasks, where the goal was the object isolation task. Agent training took place in a simulated robot learning environment with the use of primarily a coordinate-based low dimensional statespace, where reward shaping was the primary tool to teach a given task.
The reinforcement... (More) - This thesis employs deep reinforcement learning, a branch of machine learning, to carry out robotic tasks. The objective centers around teaching an agent controlling a seven-axis robot arm with a gripper tool, to complete an object isolation task. For this task, a robot manipulates a cluttered environment in such a way that a predetermined target object becomes isolated. Sub-tasks were developed to explore simpler robot tasks to evolve and combine into more complex tasks, where the goal was the object isolation task. Agent training took place in a simulated robot learning environment with the use of primarily a coordinate-based low dimensional statespace, where reward shaping was the primary tool to teach a given task.
The reinforcement learning algorithm Proximal policy optimization (PPO) implemented with a neural network architecture was used to train agents for the robotics tasks and the robot arm’s joint velocities were used as the action-space for the agents. Multiple experiments were conducted for agents practicing different tasks and their performance was evaluated by measuring their task completion rate and rendering their behavior among others. Agents developed policies capable of different forms of cube manipulation and performing cube extraction tasks. Multiple different policies for completing robot tasks were learned, and their strategies were evaluated and discussed. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9174579
- author
- Åstrand, Teodor
- supervisor
- organization
- year
- 2024
- type
- H3 - Professional qualifications (4 Years - )
- subject
- report number
- TFRT-6255
- other publication id
- 0280-5316
- language
- English
- id
- 9174579
- date added to LUP
- 2024-09-16 08:50:14
- date last changed
- 2024-09-16 08:50:14
@misc{9174579, abstract = {{This thesis employs deep reinforcement learning, a branch of machine learning, to carry out robotic tasks. The objective centers around teaching an agent controlling a seven-axis robot arm with a gripper tool, to complete an object isolation task. For this task, a robot manipulates a cluttered environment in such a way that a predetermined target object becomes isolated. Sub-tasks were developed to explore simpler robot tasks to evolve and combine into more complex tasks, where the goal was the object isolation task. Agent training took place in a simulated robot learning environment with the use of primarily a coordinate-based low dimensional statespace, where reward shaping was the primary tool to teach a given task. The reinforcement learning algorithm Proximal policy optimization (PPO) implemented with a neural network architecture was used to train agents for the robotics tasks and the robot arm’s joint velocities were used as the action-space for the agents. Multiple experiments were conducted for agents practicing different tasks and their performance was evaluated by measuring their task completion rate and rendering their behavior among others. Agents developed policies capable of different forms of cube manipulation and performing cube extraction tasks. Multiple different policies for completing robot tasks were learned, and their strategies were evaluated and discussed.}}, author = {{Åstrand, Teodor}}, language = {{eng}}, note = {{Student Paper}}, title = {{Robot Reinforcement Learning for Object Isolation}}, year = {{2024}}, }