Jun Tominaga, Maurício G.V. Ferreira and José D.S. da Silva

A rule-based satellite simulator was conceived for the verification of flight operations plans in artificial satellite control activities. Flight operations plans contain the scheduling of real-time procedures to be executed by the ground control segment, including telecommands that must be sent for the execution of mission operations by artificial satellites. Such plans, when generated by new planner software, cannot be considered entirely reliable. The safety of the missions can be ensured by evaluating the effects of scheduled telecommands on each satellite, before their actual execution. If a simulation result indicates that an unsafe internal state is reached, planning errors can be detected and corrected until an acceptable plan is obtained. The simulator consists basically on a rule-based inference engine and its associated database files. The dynamic behavior of the internal state of a satellite is defined by a set of rules. These rules are processed by the inference engine at each simulation step, in order to update the internal state parameters. The database files contain the rules that describe the system dynamics, the parameters that represent the internal state, and a queue of events, which includes the telecommands obtained from the operations plan under test and orbital events forecast by flight dynamics experts. This work describes the simulation architecture, the activities performed by the inference engine, and the data structures conceived to represent the knowledge inside the database.

Keywords

References