Abstract
Abstract
Using drones in forest fire prevention and geological surveys is becoming increasingly widespread, but it also brings safety hazards. Due to the complex forest environment, drones face issues such as low stability, long path planning, and inefficient dynamic obstacle avoidance. If a drone crashes in such an environment, it may trigger wildfires, causing enormous losses. Therefore, correct and safe drone path planning is crucial. However, current drone path planning often only considers reducing time and distance, neglecting risk costs. Hence, this paper proposes a drone path planning method based on third-party risk modeling. This approach optimizes terrain maps by considering obstacles, and forming a three-dimensional risk map. It uses ant colony algorithms to assess risks and re-implement path planning. The research shows that in planning new routes, the minimum path length is significantly lower than the average path length, reducing risks and improving path planning efficiency. This makes drone path planning more convenient and reliable. This study’s findings apply to drone path planning in high-risk areas.