Reinforcement learning-based route guidance system with dynamic traffic condition
This paper proposes a method of using reinforcement learning to solve dynamic route planning problems, and the change from static learning rate to dynamic learning rate is capable of dealing with emergent congestion. Firstly, some conventional algorithms and reinforcement learning methods are introd...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis-Master by Coursework |
| Language: | English |
| Published: |
Nanyang Technological University
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/155442 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | This paper proposes a method of using reinforcement learning to solve dynamic route planning problems, and the change from static learning rate to dynamic learning rate is capable of dealing with emergent congestion. Firstly, some conventional algorithms and reinforcement learning methods are introduced in chapter 2. Chapter 3 will discuss the software tool used for creating environment simulation and some basis of reinforcement learning. Then, the comparison of conventional reinforcement learning and proposed method is shown in detail on chapter 4. Lastly, chapter 6 is aimed at discussing some problems of proposed method and future works how to solve this problem. |
|---|