logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

A multi-objective chaotic evolution optimization framework for multi-UAV path planning in complex 3D environments

delete2026-08-17
delete0
PRE
AI
Q
Quancheng Pu
Y
Yang Lu *
T
Tieshan Li
DOI：10.1016/j.chaos.2026.118890delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
Multi-UAV path planning in complex three-dimensional environments is challenged by expanding search spaces, conflicting objectives, and the gap between discrete paths and executable trajectories. To address these issues, this paper proposes a multi-objective chaotic evolutionary optimization framework based on relative-step cylindrical-coordinate encoding. The framework constrains and decouples path variables through a relative cylindrical representation, integrates SPM-based chaotic initialization, E-DM mutation, leader-guided search, Pareto lens mapping, and grid-based archive management into the multi-objective optimization process, and converts discrete Pareto paths into smooth, time-parameterized trajectories using Chaikin subdivision, arc-length reparameterization, and Gaussian filtering. The proposed method is evaluated through encoding comparisons, component ablation studies, ZDT and WFG benchmark tests, and single- and multi-UAV path-planning experiments. The results show that the proposed encoding achieves favorable feasible-space coverage and valid-path generation, while each key component contributes positively to algorithm performance. Compared with the selected baselines, MOCEO reduces the inverted generational distance by up to 96.43% and the aggregate path cost in multi-UAV scenarios by 7.10%. It also achieves statistically significant improvements over all baselines across nine single-UAV and nine multi-UAV scenarios. Under the simulated conditions considered in this study, the post-processed trajectories satisfy the prescribed velocity and acceleration constraints. These results indicate that the proposed integrated framework effectively improves the solution quality and stability of multi-objective path planning in complex three-dimensional environments.

Journal

C
CHAOS SOLITONS & FRACTALS
IF:
5.6
Papers:
364
Citations:
0

Organization

No organization information available