Comparative Analysis of A^* and RRT^* Pathfinding Algorithms for Autonomous Drone Navigation in Various Environments

The increased utilization of drones necessitates safe and efficient traversal through various environments. This research compares the A<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">*</sup> and Rapidly-exploring Random Tree<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">*</sup> (RRT<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">*</sup>) pathfinding algorithms across four maze types: single solution, high-density, low-density, and ladder, with each maze type representing a distinct real-world environment. Four metrics were tested: flight time, flight distance, compute time, and error. A<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">*</sup> outperformed RRT<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">*</sup> in compute time for all maze types. RRT<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">*</sup> outperformed $A^{*}$ in flight time and distance for low-density, high-density, and ladder mazes, while performing similarly to A<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">*</sup> in single solution mazes. These findings can guide algorithm selection for specific situational needs.