.png){kind=icon}
.png){kind=icon}
Return
Enhanced PSO with multiple strategies for 3D UAV path planning
.png){kind=icon}
PRE
AI
X
C
H
Q
L
DOI：10.1007/s10586-026-06423-z.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
To address the challenges of unmanned aerial vehicle (UAV) path planning in complex three-dimensional (3D) environments, this paper proposes an enhanced particle swarm optimization algorithm named ACO-initialized Greedy and Escape PSO (AGAE-PSO). The proposed algorithm integrates three key strategies: an ACO-based initialization mechanism to generate high-quality initial particles, a dynamic aggressive greedy search mechanism to strengthen local exploitation, and a stagnation-aware escape mechanism to improve population diversity and avoid premature convergence. To evaluate its effectiveness, AGAE-PSO was tested on six DEM-based 3D terrain scenarios and compared with several representative optimization algorithms, including PSO, QPSO, SPSO, DE, GA, GWO, and ACOR. Experimental results demonstrate that the proposed algorithm consistently achieves superior optimization performance, producing shorter and smoother flight paths with faster and more stable convergence. In several complex scenarios, the proposed AGAE-PSO achieves approximately 20%–33% lower mean path cost than the standard PSO algorithm. In addition, AGAE-PSO achieves the best Friedman ranking value (FK = 1.00) among all compared methods while maintaining the lowest standard deviation in most scenarios, demonstrating excellent robustness and stability. These results confirm that the proposed hybrid strategy effectively balances global exploration and local exploitation for UAV path planning in complex 3D environments.
Keywords:
Unmanned aerial vehicle
Path planning
Particle swarm algorithm (PSO)
ACO-initialized Greedy and Escape PSO (AGAE-PSO)
Journal
C
IF:
4.1
Papers:
4.8K
Citations:
7.5K
Organization
Cited Papers