UR robot manipulator collision avoidance for static obstacles via path planning
As one of the most attractive machine learning technologies, deep reinforcement learning has achieved great success in many applications. The aim of this thesis is to realize the static obstacle avoidance of UR manipulator by exploring the advantages and disadvantages of deep reinforcement learni...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis-Master by Coursework |
| Language: | English |
| Published: |
Nanyang Technological University
2021
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/153122 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | As one of the most attractive machine learning technologies, deep reinforcement
learning has achieved great success in many applications. The aim of this thesis is to
realize the static obstacle avoidance of UR manipulator by exploring the advantages
and disadvantages of deep reinforcement learning and RRT and its derivation
algorithm. However, DDPG, DQN and other algorithms are difficult to achieve
obstacle avoidance, simulation results are difficult to converge, deep reinforcement
learning needs further development. Through Matlab simulation, we can realize the
static obstacle avoidance of the manipulator by RRT, RRT-connect, RRT* algorithm
and the kinematics of the manipulator. |
|---|