This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 1 with difficulty 6 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 1
Difficulty: 6
Algorithm: PPO
Episode Length: 3000
Training max_steps: 1800000
Testing max_steps: 180000
Train & Test Scripts
Download the Environment
Evaluation results
- Crash Count on hivex-aerial-wildfire-suppressionself-reported0.15000000447034836 +/- 0.20160177895130083
- Extinguishing Trees on hivex-aerial-wildfire-suppressionself-reported21.208333425223827 +/- 28.328527795039527
- Extinguishing Trees Reward on hivex-aerial-wildfire-suppressionself-reported1060.4166695594788 +/- 1416.426402849315
- Fire Out on hivex-aerial-wildfire-suppressionself-reported0.17500000223517417 +/- 0.2885484972637629
- Fire too Close to City on hivex-aerial-wildfire-suppressionself-reported0.9 +/- 0.26157418189029846
- Preparing Trees on hivex-aerial-wildfire-suppressionself-reported620.0166644275189 +/- 644.0197330961348
- Preparing Trees Reward on hivex-aerial-wildfire-suppressionself-reported620.0166644275189 +/- 644.0197330961348
- Water Drop on hivex-aerial-wildfire-suppressionself-reported48.40833265781403 +/- 37.70947155566527
- Water Pickup on hivex-aerial-wildfire-suppressionself-reported48.12500007152558 +/- 37.613333288243446
- Cumulative Reward on hivex-aerial-wildfire-suppressionself-reported1735.4933523178101 +/- 1718.2560586121042