Stratospheric Airship Path Planning Based on Improved RRT Under Constraints

Stratospheric airships are considered typical near-space low-speed vehicles characterized by large volume, low dynamics, and limited control ability. Due to environmental influences and various constraints, path planning for stratospheric airships is a highly challenging task. In addressing the issues of lengthy planning times and low efficiency in constrained conditions, we enhanced the node expansion process of the Rapidly-exploring Random Tree (RRT) algorithm. This was achieved by introducing an artificial potential field to increase sampling probabilities around the target point, guiding tree nodes to bias towards the target point for a more efficient pathfinding. The algorithm's convergence speed and sampling efficiency were notably improved. Simulation results demonstrated that our improved RRT(IRRT) algorithm reduced average planning time by approximately 70%, decreased the number of nodes by over 50 %, and significantly improved path planning efficiency compared to traditional RRT algorithms for stratospheric airship path planning.