Dynamic UAV Path Planning Under Wildfire Scenarios Using PPO Algorithm

In forest fire scenarios, unmanned aerial vehicles (UAV) must provide precise fire information for firefighting UAV. However, The dynamic and rapidly evolving nature of wildfires significantly increases the complexity of UAV-based fire reconnaissance missions. For many traditional algorithms, they are typically designed based on the assumption of UAV operations in static environments. To address this limitation, we propose an algorithm that integrates fire spread prediction with reinforcement learning (RL)-based path planning. This proposed algorithm enables adaptive trajectory optimization that dynamically responds to wildfire evolution. Finally, the simulation is carried out on the Pycharm platform to verify the effectiveness of the dynmaic scenarios.