This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 1 with difficulty 3 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 1
Difficulty: 3
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-reported0.3388888955116272 +/- 0.28207207032208675
- Extinguishing Trees on hivex-aerial-wildfire-suppressionself-reported16.488888897374274 +/- 25.98319189651514
- Extinguishing Trees Reward on hivex-aerial-wildfire-suppressionself-reported824.4444324493409 +/- 1299.1595474984952
- Fire Out on hivex-aerial-wildfire-suppressionself-reported0.1083333358168602 +/- 0.1970172357255564
- Fire too Close to City on hivex-aerial-wildfire-suppressionself-reported0.716666667163372 +/- 0.4192028074228474
- Preparing Trees on hivex-aerial-wildfire-suppressionself-reported654.4472227096558 +/- 552.5331585615204
- Preparing Trees Reward on hivex-aerial-wildfire-suppressionself-reported654.4472227096558 +/- 552.5331585615204
- Water Drop on hivex-aerial-wildfire-suppressionself-reported38.655555725097656 +/- 17.813019068968373
- Water Pickup on hivex-aerial-wildfire-suppressionself-reported38.36388869285584 +/- 17.753789101902516
- Cumulative Reward on hivex-aerial-wildfire-suppressionself-reported1660.678343963623 +/- 2522.553744525116