This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 1 with difficulty 10 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 1
Difficulty: 10
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-reported0.0833333358168602 +/- 0.13786170275335063
- Extinguishing Trees on hivex-aerial-wildfire-suppressionself-reported22.591666561365127 +/- 27.403870739917945
- Extinguishing Trees Reward on hivex-aerial-wildfire-suppressionself-reported1129.5833377838135 +/- 1370.1935526283442
- Fire Out on hivex-aerial-wildfire-suppressionself-reported0.3166666693985462 +/- 0.3819770073597531
- Fire too Close to City on hivex-aerial-wildfire-suppressionself-reported0.95 +/- 0.22360679774997894
- Preparing Trees on hivex-aerial-wildfire-suppressionself-reported520.3083327770233 +/- 466.07708625352507
- Preparing Trees Reward on hivex-aerial-wildfire-suppressionself-reported520.3083327770233 +/- 466.07708625352507
- Water Drop on hivex-aerial-wildfire-suppressionself-reported70.60000047683715 +/- 43.152979042955415
- Water Pickup on hivex-aerial-wildfire-suppressionself-reported70.2166657447815 +/- 43.11124175457642
- Cumulative Reward on hivex-aerial-wildfire-suppressionself-reported1891.3833183288575 +/- 1934.719348429914