A multiobjective giant armadillo algorithm based on two-archive for UAV path planning

Traditional UAV path planning algorithms are typically single-objective optimization methods that focus solely on minimizing the path distance. This approach neglects the impact of potential threats on the UAV’s path. To address this limitation, this paper proposes a multi-objective Armadillo optimization algorithm based on two-archive, the proposed algorithm optimizes the UAV’s path by considering both the path length and the level of threat. First, the algorithm mathematically models the equations for simulating the movement and predation location updates of the giant armadillo, thereby enhancing the search capability of the algorithm within the solution space. Second, the algorithm introduces two external archives: the Diversity Archive (DA) and the Convergence Archive (CA), which are used to store the sets of optimal solutions for diversity and convergence, respectively. Finally, the algorithm integrates the simulated binary crossover operator and the polynomial mutation operator from genetic algorithms, utilizing the CA and DA to generate leaders for each individual in the population, guiding the evolutionary process. The experimental results indicate that the proposed algorithm exhibits strong competitiveness in the domain of UAV path planning.