This model serves as the baseline for the Aerial Wildfire Suppression environment, trained and tested on task 3 with difficulty 1 using the Proximal Policy Optimization (PPO) algorithm.
Environment: Aerial Wildfire Suppression
Task: 3
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-reported0.14166667088866233 +/- 0.18945135260751206
- Extinguishing Trees on hivex-aerial-wildfire-suppressionself-reported1.041666679084301 +/- 3.059772488050604
- Extinguishing Trees Reward on hivex-aerial-wildfire-suppressionself-reported5.208333283662796 +/- 15.298862328050262
- Fire Out on hivex-aerial-wildfire-suppressionself-reported0.4916666686534882 +/- 0.427405472920952
- Fire too Close to City on hivex-aerial-wildfire-suppressionself-reported0.575 +/- 0.46665100224336487
- Preparing Trees on hivex-aerial-wildfire-suppressionself-reported654.5333339691163 +/- 369.49324566338055
- Preparing Trees Reward on hivex-aerial-wildfire-suppressionself-reported654.5333339691163 +/- 369.49324566338055
- Water Drop on hivex-aerial-wildfire-suppressionself-reported19.52499990463257 +/- 10.494119202517384
- Water Pickup on hivex-aerial-wildfire-suppressionself-reported19.108333349227905 +/- 10.481433832592886
- Cumulative Reward on hivex-aerial-wildfire-suppressionself-reported722.9217496871948 +/- 414.79654723295414