Six-Degree of Freedom Twin-Jet Aircraft Simulation

With the stringent requirements placed on current autoland systems for commercial aircraft, simulation has become an essential tool in the certification of these autoland systems. Current requirements established by the Federal Aviation Administration (FAA) and Joint Aviation Administrations (JAA) mandate that the probability of an undetected failure which could cause a catastrophic event must be 10-9 per approach and the probability of an undetected failure which could cause a hazardous but not catastrophic event must be less than 10-7 per approach. In order to demonstrate compliance with these requirements, aircraft manufacturers use software simulations that simulate the performance of any signal used as guidance, the airborne sensor, the autoland system, the aircraft, and the physical environment. These simulations are run in monte-carlo iterations where the probability of certain types of failures are assumed. The overall performance of the system can then be evaluated to the required statistical precision based on the simulation results. During these simulations, it is assumed that the weather is always bad, that is winds are set at the standard models used by the FAA or JAA.

The simulation that I will discuss has been used by the National Aeronautics and Space Administration (NASA) to simulate the performance of their twin-jet engine test aircraft. This simulation was written completely in FORTRAN and was undocumented. The purpose of this work was to implement the simulation in the MATLAB SIMULINK environment and document the code so that it will be more flexible and more easily modified. A primary advantage of the SIMULINK environment is that a variety of tools are available to aid controller design and evaluate system performance. These tools enable a user to trim the model about some operating point, find linearized models of systems, and conveniently display results or store them in variable for further analysis in MATLAB.