Robot obstacle avoidance using reinforcement learning

Path planning is one of the essential parts of the autonomous robotic field and cars. There are many paths planning algorithms use to solve the static environment. However, path planning for a dynamic environment is challenging in this robotic field. Some traditional path planning strategies pre-def...

Full description

Saved in:
Bibliographic Details
Main Author: Nang Yune Thitsar
Other Authors: Hu Guoqiang
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2022
Subjects:
Online Access:https://hdl.handle.net/10356/158511
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Path planning is one of the essential parts of the autonomous robotic field and cars. There are many paths planning algorithms use to solve the static environment. However, path planning for a dynamic environment is challenging in this robotic field. Some traditional path planning strategies pre-defined the robots' routes since the climate is already known. To face the dynamic environment, robots need to be more intelligent. Reinforcement learning is one of the machine learning techniques used to solve the dynamic environment and complex situations.