Stratified sampled equilibrium optimizer for agile earth observation satellite observation tasks scheduling

Metaheuristic algorithms constitute one of the most efficient methods for addressing the Agile Earth Observation Satellite Scheduling Problem (AEOSSP). Hence, exploring the capability of a new metaheuristic algorithm called Equilibrium Optimizer (EO) in this domain, is one of the key contributions o...

Full description

Saved in:
Bibliographic Details
Published in:Computing Vol. 107; no. 12; p. 233
Main Authors: Galloua, Mohamed Elamine, Li, Shuai, Cui, Jiahao
Format: Journal Article
Language:English
Published: Vienna Springer Vienna 01.12.2025
Springer Nature B.V
Subjects:
Algorithms
Artificial Intelligence
Coding
Computer Appl. in Administrative Data Processing
Computer Communication Networks
Computer Science
Earth observations (from space)
Equilibrium
Evolution & development
Flight
Genetic algorithms
Heuristic methods
Information Systems Applications (incl.Internet)
Integers
Optimization algorithms
Regular Paper
Representations
Sampling
Satellite observation
Satellites
Schedules
Scheduling
Software Engineering
Stagnation
Unmanned aerial vehicles
Stratified sampling
Agile earth observation satellite
Equilibrium optimizer
Tasks scheduling
68T20
90B36
90C27
Lévy flight
ISSN:0010-485X, 1436-5057
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Metaheuristic algorithms constitute one of the most efficient methods for addressing the Agile Earth Observation Satellite Scheduling Problem (AEOSSP). Hence, exploring the capability of a new metaheuristic algorithm called Equilibrium Optimizer (EO) in this domain, is one of the key contributions of this paper. In this work we aim to analyzes the limitations of the original EO algorithm in addressing the specific requirements of AEOSSP. Specifically, the integer-point data representation within the algorithm population, emerges as a challenge for this subject. In response, a new variant, Stratified Sampling Equilibrium Optimizer (SSEO) is proposed to address these deficiencies. Taking into account the specificity of this field, targeted improvements have been introduced. First, the EO algorithm was adapted to support integer-point representation, ensuring compatibility with the discrete encoding scheme. Second, a stratified sampling mechanism is incorporated to mitigate variable duplication across particles, a critical requirement since the AEOSSP formulation strictly prohibits task repetition within a single schedule. Finally, a stagnation mitigation strategy is introduced, leveraging a tailored Lévy flight mechanism. This mechanism includes a customized Lévy step generation method, designed specifically to align with the characteristics of the AEOSSP. The proposed SSEO algorithm is validated using real-world satellite data and demonstrates superior performance across multiple aspects, including solution quality (fitness), computational efficiency (processing time), and robustness against local optima. The implementation code for the proposed Stratified Sampling Equilibrium Optimizer (SSEO) algorithm is publicly available at: https://github.com/EnimaG/SSEO . Highlights The EO algorithm is modified to support integer-point representation, enabling effective handling of AEOSSP’s discrete encoding requirements. A stratified sampling mechanism is implemented to prevent task duplication across particles, ensuring feasible schedules and improving population diversity. A tailored Lévy flight-based mechanism with adaptive step generation is introduced to mitigate stagnation and enhance exploration.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0010-485X
1436-5057
DOI:10.1007/s00607-025-01588-8