Intelligent decision support system for Manned Spaceflight Launching

In order to reduce the decision fault of commander, an intelligent decision support system for Manned Spaceflight Launching was developed. The configuration, functions and inference process of the system were described in detail. The inference method of relation matrices, decision trees and fault dictionaries were mainly discussed. The input information corresponding to the output information one by one was used by the relation matrices to establish the mapping relation between the fault mode and test information to diagnose and locate the faults. The minimal launching conditions are determined by the decision trees according to three tiers of leaf, judgment and decision nodes. Fault dictionaries are realized by the fuzzy fault dictionaries step by step, in which the fault dictionaries are set up both on the level of system, subsystems and equipments and on the level of vehicles, ground equipments and artificial information. Afterwards, the policy to dealing with faults can be provided rapidly by the fuzzy inference method. The system was applied to Manned Spaceflight Launching successfully, which verify that the real-time characteristic of the system and inference result met the requirement of launching commander.