Abstract

Evolutionary algorithms, such as particle swarm optimization (PSO), are widely applied to UAV path planning problems. However, the fixed particle length of PSO, which may not be suitable for the scenario, will compromise the search efficiency. This paper proposes the RGG-PSO+ method, which adapts to scenarios by dynamically adjusting the number of waypoints. Random geometric graphs (RGG) and the divide-and-conquer paradigm are involved in improving the proposed method. Comparative analyses with established heuristic methods demonstrate RGG-PSO+’s superior performance in complex environments, particularly in terms of convergence speed and path length. The implementation of RGG significantly improves the F-Measure, indicating a shift from exploration to exploitation of PSO’s iterations, and the implementation of the divide-and-conquer paradigm is evident in the improved mean and variance of normalized path lengths.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call