Autonomous Exploration Path Planning Method for Unmanned Aerial Vehicle Based on Flight Safety Constraint

When UAVs explore autonomously in indoor environments, they need to keep a safe distance from the surrounding environment to ensure flight safety. In view of this, this paper proposes an autonomous exploration path planning method for UAVs based on flight safety constraints: firstly, an improved RRG passable path generation algorithm based on voxel collision detection is proposed to construct a 3D raster map by LiDAR and optimize RRG nodes based on voxel collision detection, which achieves a safe sampling and connecting of RRG nodes of the global paths; followed by an obstacle avoidance constraint-based improved A* path planning algorithm is proposed based on obstacle avoidance constraints, and the obstacle avoidance penalty function is added to improve the safety of the local paths; finally, the paths are optimized based on the idea of breadth-first to shorten the path lengths and improve the smoothness. The proposed algorithm is verified by gazebo simulation platform, which shows that the path planned in this paper is safer than the traditional algorithm, and can effectively reduce the path length and planning time in the process of autonomous exploration.