Research On Task Scheduling Based On Earth Observation Satellite

With the development of satellite communication technology and information processing technology,earth observation satellite(EOS)has been widely used in land surveying and mapping,natural disaster monitoring,military reconnaissance and other fields due to their wide coverage,long duration,and freedom from national boundaries and time domain restrictions.In the process of observing the earth by satellite,the numerous earth observation tasks and limited satellite resources cause problems such as high task conflict and low execution efficiency,which brings huge challenges to the implementation of satellite tasks.Based on the EOS network,this thesis conducts research on EOS-based task scheduling by focusing on the low task execution efficiency,low benefit and low satellite resource utilization in EOS.The main contents are summarized as follows:(1)The observation task scheduling scheme based on tabu genetic algorithm is proposed.Aiming at problems such as low scheduling efficiency and low benefit caused by the surge and unbalanced distribution of user earth observation tasks,an observation task scheduling scheme based on layered tabu genetic algorithm(LTGA)is proposed.Firstly,an EOS observation network model consisting of the EOS and ground stations is constructed.Then,in order to reduce task conflict and improve the efficiency of task scheduling,the task splitting thresholds and clustering thresholds are defined.The tasks are divided into common task set and superior task set,after the tasks are preprocessed.Finally,by introducing LTGA,based on two sets of tasks successively,tasks are selected and sorted.As a result,the optimal task scheduling plan is obtained.The simulation results show that the proposed observation task scheduling scheme can effectively improve the scheduling rewards,and the effect becomes higher as the number of tasks increases.(2)The transmission task scheduling scheme based on the connection cost-effective is proposed.Aiming at the problems of low task execution immediacy,low satellite resources utilization,and low task execution flexibility caused by limited storage,limited energy and fixed channel status,a transmission task scheduling scheme based on the best connection efficiency is proposed.Firstly,an EOS transmission network model consisting of the EOS,relay satellites,communication satellites,and ground stations is constructed.Then,a time expansion graph based on link distribution and channel quality is proposed.At the same time,comprehensively considering the relationship between the link status and node status,the connection state weight factor is defined.Finally,in order to maximize cost-effectiveness,the EOS-based transmission task scheduling problem is modeled as a mixed integer nonlinear problem,and the problem is solved by equivalently splitting the problem into a link connection problem and a task scheduling problem.The simulation results show that the proposed transmission task scheduling scheme can improve the immediacy of task completion and the utilization of satellite resources.