Abstract
This paper discusses the implementation of a Deep
Reinforcement Learning policy, based on DQN, which optimizes
the navigation of the UAV to the front of wind turbine blades.
The UAV was trained in simulation using Unreal Engine V4.27
coupled with AirSim. The action space of the UAV was discretized
while allowing 6 different actions to be executed. A Yolov5
network trained with images of simulated wind turbines was
used for detection and tracking, providing the DQN policy with
state information, upon which it has been trained. In addition to
this, the dynamic reward has been implemented, which combined
both navigation and inspection objectives in the final evaluation
of actions. Our tests showed that after 7500 time-steps the
exploration rate reached near 0, the mean length of the episodes
increased from 10 down to 30, but the mean reward increased
from around -60 to stabilizing the output at 26. These results
suggest that the proposed method is a promising solution to
optimizing the autonomous inspection of wind turbines with
UAVs.
Reinforcement Learning policy, based on DQN, which optimizes
the navigation of the UAV to the front of wind turbine blades.
The UAV was trained in simulation using Unreal Engine V4.27
coupled with AirSim. The action space of the UAV was discretized
while allowing 6 different actions to be executed. A Yolov5
network trained with images of simulated wind turbines was
used for detection and tracking, providing the DQN policy with
state information, upon which it has been trained. In addition to
this, the dynamic reward has been implemented, which combined
both navigation and inspection objectives in the final evaluation
of actions. Our tests showed that after 7500 time-steps the
exploration rate reached near 0, the mean length of the episodes
increased from 10 down to 30, but the mean reward increased
from around -60 to stabilizing the output at 26. These results
suggest that the proposed method is a promising solution to
optimizing the autonomous inspection of wind turbines with
UAVs.
| Original language | English |
|---|---|
| Title of host publication | 9th 2023 International Conference on Control, Decision and Information Technologies, CoDIT 2023 |
| Number of pages | 6 |
| Publisher | IEEE (Institute of Electrical and Electronics Engineers) |
| Publication date | Oct 2023 |
| Pages | 2372-2377 |
| Article number | 10284087 |
| ISBN (Print) | 979-8-3503-1141-9 |
| ISBN (Electronic) | 979-8-3503-1140-2 |
| DOIs | |
| Publication status | Published - Oct 2023 |
| Event | 9th International Conference on Control, Decision and Information Technologies (CoDIT) - Rome, Italy Duration: 3 Jul 2023 → 6 Jul 2023 https://codit2023.com/ |
Conference
| Conference | 9th International Conference on Control, Decision and Information Technologies (CoDIT) |
|---|---|
| Country/Territory | Italy |
| City | Rome |
| Period | 03/07/2023 → 06/07/2023 |
| Internet address |
| Series | International Conference on Control, Decision and Information Technologies (CoDIT) |
|---|---|
| ISSN | 2576-3555 |
Keywords
- Condition Monitoring
- Deep Q-network
- Deep Reinforcement Learning
- Dynamic Reward
- Inspection
- Path-planning
- Simulation
- UAV