This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 4 with difficulty 4 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 4
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-reported0.995833334326744 +/- 0.01863389536983029
- Extinguishing Trees on hivex-aerial-wildfire-suppressionself-reported0.0886829849332571 +/- 0.2431451047770778
- Extinguishing Trees Reward on hivex-aerial-wildfire-suppressionself-reported0.44341491162776947 +/- 1.2157254834823454
- Fire too Close to City on hivex-aerial-wildfire-suppressionself-reported0.02242424301803112 +/- 0.04967780915490442
- Preparing Trees on hivex-aerial-wildfire-suppressionself-reported253.21703186035157 +/- 35.590698365756126
- Preparing Trees Reward on hivex-aerial-wildfire-suppressionself-reported253.21703186035157 +/- 35.590698365756126
- Water Drop on hivex-aerial-wildfire-suppressionself-reported1.6389704704284669 +/- 0.33632718660720196
- Water Pickup on hivex-aerial-wildfire-suppressionself-reported1.6389704704284669 +/- 0.33632718660720196
- Cumulative Reward on hivex-aerial-wildfire-suppressionself-reported145.59471988677979 +/- 48.609159328951485