Litcius/Paper detail

Autonomous Task Planning Method for Multi-Satellite System Based on a Hybrid Genetic Algorithm

Jun Long, Shimin Wu, Xiaodong Han, Yunbo Wang, Limin Liu

2023Aerospace33 citationsDOIOpen Access PDF

Abstract

The increasing number of satellites for specific space tasks makes it difficult for traditional satellite task planning that relies on ground station planning and on-board execution to fully exploit the overall effectiveness of satellites. Meanwhile, the complex and changeable environment in space also poses challenges to the management of multi-satellite systems (MSS). To address the above issues, this paper formulates a mixed integer optimization problem to solve the autonomous task planning for MSS. First, we constructed a multi-agent-based on-board autonomous management and multi-satellite collaboration architecture. Based on this architecture, we propose a hybrid genetic algorithm with simulated annealing (H-GASA) to solve the multi-satellite cooperative autonomous task planning (MSCATP). With the H-GASA, a heuristic task scheduling scheme was developed to deal with possible task conflicts in MSCATP. Finally, a simulation scenario was established to validate our proposed H-GASA, which exhibits a superior performance in terms of computational power and success rate compared to existing algorithms.

Topics & Concepts

Computer scienceTask (project management)Simulated annealingGenetic algorithmScheduling (production processes)SatelliteHeuristicArchitectureDistributed computingExploitReal-time computingArtificial intelligenceAlgorithmSystems engineeringMathematical optimizationMachine learningEngineeringVisual artsArtMathematicsComputer securityAerospace engineeringSatellite Communication SystemsMobile Agent-Based Network ManagementOptimization and Search Problems
Autonomous Task Planning Method for Multi-Satellite System Based on a Hybrid Genetic Algorithm | Litcius