Zexuan Pei
Zexuan Pei has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Jichun Li
Newcastle University
1 shared publication - Jiguang Li
Department of Computer and Information Science, Northumbria University at Newcastle, Newcastle, U.K.
1 shared publication - Lei Bai
School of Electrical Engineering, North China University of Water Resources and Electric Power, Zhengzhou 450045, China
1 shared publication - Xing Liu1 shared publication
- Yu Zhou
School of Electrical Engineering, North China University of Water Resources and Electric Power, Zhengzhou, China
1 shared publication
Latest Publications
- Open Access
- 2 February 2026
- Published by Elsevier in SSRN Electronic Journal
Solar Activity Optimization Algorithm for Multi-UAV Path Planning
Multi-UAV collaborative path planning is a core challenge in autonomous navigation, involving the search for optimal or near-optimal paths under multiple constraints. this paper proposes a novel meta-heuristic algorithm, termed the Solar Activity Optimizer (SAO), SAO models three types of solar activities: (1) adaptive exploration, achieved through a random screening strategy guided by the periodic variation of sunspots to balance global search ; (2) energy outburst, realized via a flare-triggered directional mutation mechanism that enhances the ability to escape local optima; (3) local exploitation, implemented using a logarithmic spiral trajectory to improve local search accuracy. A center-offset experiment further confirms that the observed performance improvements arise from the intrinsic algorithmic design rather than benchmark-specific characteristics. multi-UAV path planning experiments involving five UAVs are conducted across four complex scenarios constructed from digital elevation model (DEM) data of two real mountainous regions. The results show SAO can generate safe, smooth flight paths with reduced collision risk, validating its effectiveness and practical applicability in complex multi-UAV path planning tasks, ablation studies based on migrating SAO’s core strategies into PSO confirm both the random screening strategy and the partial individual mutation strategy play critical roles in improving the safety and robustness of path planning.
- 2 February 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSolar Activity Optimization Algorithm for Multi-UAV Path Planning
Multi-UAV collaborative path planning is a core challenge in autonomous navigation, involving the search for optimal or near-optimal paths under multiple constraints. this paper proposes a novel meta-heuristic algorithm, termed the Solar Activity Optimizer (SAO), SAO models three types of solar activities: (1) adaptive exploration, achieved through a random screening strategy guided by the periodic variation of sunspots to balance global search ; (2) energy outburst, realized via a flare-triggered directional mutation mechanism that enhances the ability to escape local optima; (3) local exploitation, implemented using a logarithmic spiral trajectory to improve local search accuracy. A center-offset experiment further confirms that the observed performance improvements arise from the intrinsic algorithmic design rather than benchmark-specific characteristics. multi-UAV path planning experiments involving five UAVs are conducted across four complex scenarios constructed from digital elevation model (DEM) data of two real mountainous regions. The results show SAO can generate safe, smooth flight paths with reduced collision risk, validating its effectiveness and practical applicability in complex multi-UAV path planning tasks, ablation studies based on migrating SAO’s core strategies into PSO confirm both the random screening strategy and the partial individual mutation strategy play critical roles in improving the safety and robustness of path planning.