Parallel Distributional Deep Reinforcement Learning for Mapless Navigation of Terrestrial Mobile Robots
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912009794093056 |
|---|---|
| author | Kich, Victor Augusto Kolling, Alisson Henrique de Jesus, Junior Costa Heisler, Gabriel V. Jacobs, Hiago Bottega, Jair Augusto Kelbouscas, André L. da S. Ohya, Akihisa Grando, Ricardo Bedin Drews-Jr, Paulo Lilles Jorge Gamarra, Daniel Fernando Tello |
| author_facet | Kich, Victor Augusto Kolling, Alisson Henrique de Jesus, Junior Costa Heisler, Gabriel V. Jacobs, Hiago Bottega, Jair Augusto Kelbouscas, André L. da S. Ohya, Akihisa Grando, Ricardo Bedin Drews-Jr, Paulo Lilles Jorge Gamarra, Daniel Fernando Tello |
| contents | This paper introduces novel deep reinforcement learning (Deep-RL) techniques using parallel distributional actor-critic networks for navigating terrestrial mobile robots. Our approaches use laser range findings, relative distance, and angle to the target to guide the robot. We trained agents in the Gazebo simulator and deployed them in real scenarios. Results show that parallel distributional Deep-RL algorithms enhance decision-making and outperform non-distributional and behavior-based approaches in navigation and spatial generalization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_05744 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Parallel Distributional Deep Reinforcement Learning for Mapless Navigation of Terrestrial Mobile Robots Kich, Victor Augusto Kolling, Alisson Henrique de Jesus, Junior Costa Heisler, Gabriel V. Jacobs, Hiago Bottega, Jair Augusto Kelbouscas, André L. da S. Ohya, Akihisa Grando, Ricardo Bedin Drews-Jr, Paulo Lilles Jorge Gamarra, Daniel Fernando Tello Robotics This paper introduces novel deep reinforcement learning (Deep-RL) techniques using parallel distributional actor-critic networks for navigating terrestrial mobile robots. Our approaches use laser range findings, relative distance, and angle to the target to guide the robot. We trained agents in the Gazebo simulator and deployed them in real scenarios. Results show that parallel distributional Deep-RL algorithms enhance decision-making and outperform non-distributional and behavior-based approaches in navigation and spatial generalization. |
| title | Parallel Distributional Deep Reinforcement Learning for Mapless Navigation of Terrestrial Mobile Robots |
| topic | Robotics |
| url | https://arxiv.org/abs/2408.05744 |