This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 1 with difficulty 7 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 1
Difficulty: 7
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-reported0.42777778655290605 +/- 0.26862100733560945
- Extinguishing Trees on hivex-aerial-wildfire-suppressionself-reported33.80000016689301 +/- 38.84839322135217
- Extinguishing Trees Reward on hivex-aerial-wildfire-suppressionself-reported1690.0000106811524 +/- 1942.4196808320014
- Fire Out on hivex-aerial-wildfire-suppressionself-reported0.06666666865348816 +/- 0.13679711768822556
- Fire too Close to City on hivex-aerial-wildfire-suppressionself-reported0.9583333343267441 +/- 0.13106625028248262
- Preparing Trees on hivex-aerial-wildfire-suppressionself-reported636.8777788162231 +/- 568.0216883131019
- Preparing Trees Reward on hivex-aerial-wildfire-suppressionself-reported636.8777788162231 +/- 568.0216883131019
- Water Drop on hivex-aerial-wildfire-suppressionself-reported38.541666984558105 +/- 15.550133835890938
- Water Pickup on hivex-aerial-wildfire-suppressionself-reported38.38333382606506 +/- 15.581525306646949
- Cumulative Reward on hivex-aerial-wildfire-suppressionself-reported2319.882499694824 +/- 1935.6842099941402