Multiobjective 3D path planning for UAVs based on an improved adaptive artificial fish swarm algorithm

Abstract

To address the issues of slow convergence and susceptibility to local optima when applying traditional artificial fish swarm algorithms to 3D path planning for unmanned aerial vehicles (UAVs), this paper proposes an improved adaptive artificial fish swarm algorithm (IAFSA). A simulation environment incorporating undulating terrain and spherical no-fly zones was constructed based on real-world flight scenarios, and a multi-objective evaluation model was developed focusing on flight range, flight safety, and energy consumption under different flight conditions. During the iterative process, the algorithm dynamically adjusts the search horizon and step size while introducing an elite retention mechanism to preserve highquality path information. The study employs cubic spline interpolation to smooth discrete waypoints and performs subsequent safety checks. Under identical conditions, the proposed algorithm is compared with the standard artificial fish school algorithm and the particle swarm optimization algorithm through simulation. Experimental results demonstrate that the proposed algorithm converges faster, achieves a higher success rate, and generates flight paths with shorter distances, lower energy consumption, and smoother flight attitudes, thereby better accommodating UAV flight missions in complex 3D environments.