یک مدل، یک سیستم اکتشافی و یک سیستم پشتیبانی تصمیم گیری برای حل مشکل برنامه ریزی در رصد ماهواره ای صورت فلکی های زمین

hina plans to launch four small optical satellites and four small SAR satellites to form a natural disaster monitoring constellation. Data can be obtained by the constellation in all weather conditions for disaster alert and environmental damage analysis. The scheduling problem for the constellation consists of selecting and timetabling the observation activities to acquire the requested images of the earth surface and scheduling the download activities to transmit the image files to a set of ground stations. The scheduling problem is required to be solved every day in a typical 1-day horizon and it must respect complex satellite operational constraints as well as request preferences, such as visibility time windows, transition time between consecutive observations or downloads, memory capacity, energy capacity, polygon target requests and priorities. The objective is to maximize the rewards of the images taken and transmitted. We present a nonlinear model of the scheduling problem, develop a priority-based heuristic with conflict-avoided, limited backtracking and download-as-needed features, which produces satisfactory feasible plans in a very short time. A decision support system based on the model and the heuristic is also provided. The system performance shows a significant improvement with respect to faster and better scheduling of an earth observing satellite constellation.

The first environmental and disaster monitoring and prediction satellite constellation of China consists of four small optical satellites and four small SAR (Synthetic Aperture Radar) satellites, which are typical earth observing satellites. The constellation’s aim is to obtain earth images in all weather conditions and over a large area, providing the abilities for natural disaster alert and environmental damage analysis. The first phase of the project sees the launch of two optical satellites, i.e., HJ-1A and HJ-1B (HJ is short for Huan Jing, which means Environment), which were already shot in September 2008, and the first SAR satellite, HJ-1C, is planned for 2010. The two optical satellites of the constellation are located in the same orbital plane of 97.95° at 650 km altitude and are phased at 180° from each other, covering all Chinese territory in 48 h. HJ-1C will follow a heliosynchronous low orbit (about 500 km above the earth surface) around the earth at the inclination of 97.37°, and it will run 15.2258 orbits a day. The three satellites of the constellation’s first phase have complementary advantages: they will allow periodically repeated observations of a same site, and the infra-red camera of HJ-1B and the SAR technology on-board HJ-1C will provide high resolution images regardless of illumination or weather conditions. The constellation is intended for both national and international use. Besides being applied to natural disaster monitoring purpose within Chinese territory, the constellation will also make contribution to the International Charter ‘Space and Major Disasters’, a joint initiative that aims to provide emergency response satellite data free of charge to those affected by disasters anywhere in the world. The constellation is supposed to satisfy different customers, such as China National Disaster Deduction Commission, normal civilians who require for images of area interested, and China Center for Resource Satellite Data and Applications (CRESDA), who is responsible for the ground segment that consists of a Mission Command and Control Center, and several telemetry, tracking and communication ground stations both for uploading the operational commands to the satellites and for receiving images from them. We have been participating in the constellation program since January 2008 with CRESDA, who is in charge of the development of the scheduling algorithm.

This paper introduced the scheduling problem for the first environmental and disaster monitoring and prediction satellite constellation of China. A mathematical model, a heuristic algorithm and a Decision Support System were provided. The research demonstrated that analyzing and modeling specific process characteristics and supporting operations research methodology may be successfully applied to develop solution methodology and decision support systems in the earth observing satellite constellation scheduling problem domain. Although in this paper the Chinese satellite constellation was taken as a study-case, the description and the modeling method of the characteristics of the earth observing satellite scheduling domain, which are confronted by most researchers and engineers in the field, could be an option for them to formulate their own problem. Since observation and data-download are natural and inseparable two parts of the image acquisition of the earth observing satellite, it is reasonable to consider the two parts together and thus formulate one monolithic model as in this paper. However, exact methods may be inappropriate and unable to solve such model directly, and heuristic is the first choice. We presented three important points as follows when devising an efficient heuristic for a generalized earth observing satellite constellation scheduling problem: •schedule observations firstly, when memory capacity is reached, schedule downloads as needed, •focus on the disjunctive resource, precisely, focus on the observation time, and introduce some criteria to evaluate the contentions on the observation time, •in order to mitigate the possible low quality of the solution provided by an construction heuristic, incorporate some look-ahead and backtracking mechanism as long as time permits. Finally, we want to note that the model and the heuristic as well as the system are now under test-run. Since September 2008, we have been evaluating the system with real-world scenarios, which can at most solve problem instances involving three satellites, five ground stations and two hundreds requests and can take seven days as the schedule horizon. Currently the fine tuning of the heuristic is conducted. Possible future directions could be to devise problem-specific repair heuristics to improve the feasible schedule when the time requirement permits. In order to better evaluate the performance of our algorithm, some methods which can provide strict upper and lower bounds are also possible research prospects.