The corresponding corrected method is proposed for the INS (INS-Inertial Navigation System) accumulated error of large transport aircraft.System errors contain aircraft position error,altitude error and speed error,one is increasing the accuracy of hardware; the other is development of low cost software algorithms.Because of improving hardware is more difficult in my country at present,developing software algorithms is essential way,which have been validated in my types of airplane.The combined heuristic algorithms (ABPNN,Advanced Back-propagation neural networks algorithm and LSM-least square method) are presented,which incorporates the effects of flight region and measured terrain height data by radar and barometer.Based on this algorithm,the appropriate match region was gotten by recognition of fiducial digital map in real time online.In process of work,the minimum of position error as a cost function and the constraint conditions are gave,the flight positions are recognized in real time and continuously,least sum of square is calculated based on LSM,in other words,the optimized result is obtained.The simulation case demonstrate that the method is very successful,the correct rate of recognition is more 90 percent.In words,the algorithm presented is economical,validation and effective.
展开▼
