This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 0 with difficulty 4 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 0
Difficulty: 4
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-reported0.26666667461395266 +/- 0.24423170630035398
- Extinguishing Trees on hivex-aerial-wildfire-suppressionself-reported24.53333340883255 +/- 44.947390538539814
- Extinguishing Trees Reward on hivex-aerial-wildfire-suppressionself-reported122.66666712760926 +/- 224.73694526974205
- Fire Out on hivex-aerial-wildfire-suppressionself-reported0.14166666939854622 +/- 0.24348229211157768
- Fire too Close to City on hivex-aerial-wildfire-suppressionself-reported0.975 +/- 0.11180339887498947
- Preparing Trees on hivex-aerial-wildfire-suppressionself-reported818.2583312273025 +/- 623.9383921347651
- Preparing Trees Reward on hivex-aerial-wildfire-suppressionself-reported818.2583312273025 +/- 623.9383921347651
- Water Drop on hivex-aerial-wildfire-suppressionself-reported12.241666674613953 +/- 5.139423095979583
- Water Pickup on hivex-aerial-wildfire-suppressionself-reported11.858333373069764 +/- 5.146473016503177
- Cumulative Reward on hivex-aerial-wildfire-suppressionself-reported1014.383337020874 +/- 634.9201357813032