This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 1 with difficulty 1 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 1
Difficulty: 1
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-reported0.10555555783212185 +/- 0.16411299281130134
- Extinguishing Trees on hivex-aerial-wildfire-suppressionself-reported5.788888883590698 +/- 10.994661682099375
- Extinguishing Trees Reward on hivex-aerial-wildfire-suppressionself-reported289.44444465637207 +/- 549.7330871286864
- Fire Out on hivex-aerial-wildfire-suppressionself-reported0.36388889104127886 +/- 0.42710131135604623
- Fire too Close to City on hivex-aerial-wildfire-suppressionself-reported0.7083333343267441 +/- 0.40780040400115103
- Preparing Trees on hivex-aerial-wildfire-suppressionself-reported556.8888899803162 +/- 429.9611601706457
- Preparing Trees Reward on hivex-aerial-wildfire-suppressionself-reported556.8888899803162 +/- 429.9611601706457
- Water Drop on hivex-aerial-wildfire-suppressionself-reported15.513888931274414 +/- 8.362694438094918
- Water Pickup on hivex-aerial-wildfire-suppressionself-reported14.897222208976746 +/- 8.362974150430102
- Cumulative Reward on hivex-aerial-wildfire-suppressionself-reported817.7808405578137 +/- 571.7617652663462