An Improved Informed RRT*-connect Algorithm for UAVs Path Planning

With the widespread application of UAVs, the demand for efficient and accurate path planning algorithms has become increasingly important. This paper proposes an algorithm called Improved Informed RRT*-connect, which makes three key enhancements: bias extension, probabilistic guidance, and node clipping. These enhancements aim to optimize the algorithm's efficiency and path quality in complex environments. First, the bias extension directs the random tree growth towards the goal point; secondly, the probabilistic guidance strategy attempts to connect directly to the goal point with a certain probability after generating nodes; finally, node clipping optimization simplifies the path by removing unnecessary nodes. Preliminary experiments have been conducted to determine the optimal values for the bias extension weight parameter and the probabilistic guidance parameter. Moreover, the improved algorithm has been compared with several classical RRT algorithms in terms of cost time, path length, and node count. Simulation results show that in all four scenarios, the improved algorithm outperforms classical algorithms in terms of both search speed and path optimization performance.