Three-dimensional path planning for UAVs using a quantum local search-enhanced differential evolution algorithm
TL;DR
Experimental results demonstrate that QLSDE exhibits clear advantages in most scenarios for UAV path planning problems, and several classical and state-of-the-art metaheuristic algorithms were compared to validate the algorithm's significant optimization capabilities.
Abstract
This paper introduces a Quantum Local Search Differential Evolution Algorithm (QLSDE) to address path planning for unmanned aerial vehicles (UAVs) in sophisticated environments with multiple threats. First, the path planning problem is transformed into an optimization model by constructing a cost function that incorporates operational requirements and constraints, including UAV feasibility and safety. Subsequently, the QLSDE algorithm efficiently explores the configuration space by leveraging the mapping relationship between particle positions and UAV parameters (velocity, turn angle, and climb/descent angle) to minimize the cost function, thereby deriving the optimal flight path. To evaluate QLSDE's optimization performance, the present paper compared it with several classical and state-of-the-art metaheuristic algorithms (including DE, PSO, GWO, and SaUSDE). Results validated the algorithm's significant optimization capabilities. Furthermore, four benchmark test scenarios were constructed based on real digital elevation model maps. Experimental results demonstrate that QLSDE exhibits clear advantages in most scenarios for UAV path planning problems.