This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 0 with difficulty 5 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 0
Difficulty: 5
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-reported0.10833333656191826 +/- 0.13545462085140367
- Extinguishing Trees on hivex-aerial-wildfire-suppressionself-reported21.558333230018615 +/- 22.116535318538975
- Extinguishing Trees Reward on hivex-aerial-wildfire-suppressionself-reported107.79166641235352 +/- 110.58267924973548
- Fire Out on hivex-aerial-wildfire-suppressionself-reported0.3250000037252903 +/- 0.352414398829495
- Fire too Close to City on hivex-aerial-wildfire-suppressionself-reported0.975 +/- 0.11180339887498947
- Preparing Trees on hivex-aerial-wildfire-suppressionself-reported906.383334851265 +/- 686.9466761748455
- Preparing Trees Reward on hivex-aerial-wildfire-suppressionself-reported906.383334851265 +/- 686.9466761748455
- Water Drop on hivex-aerial-wildfire-suppressionself-reported42.29166712760925 +/- 14.596310384142566
- Water Pickup on hivex-aerial-wildfire-suppressionself-reported41.733333349227905 +/- 14.532822126312558
- Cumulative Reward on hivex-aerial-wildfire-suppressionself-reported1040.6658332824707 +/- 651.2068293319696