Abstract:
In recent years, more and more light aircraft enter our daily life, from Agricultural
applications, emergency rescue, flight experiment and training to Barriers to
entry, light aircraft always have their own advantages. Thus, they have become
more and more popular.
However, in the process of GDP research about Flight Control System design
for the Flying Crane, the author read a lot of literature about Flight Control
System design, then noticed that the research in Flight Control System have
apparently neglected to Low-cost vehicles. So it is necessary to do some study
about Flight Control System for this kind of airplane. The study will more
concern the control law design for ultra-light aircraft, the author hopes that with
an ‘intelligence’ Flight Control System design, this kind of aircraft could
sometimes perform flying tasks according to a prearranged flight path and
without a pilot.
As the Piper J-3 cub is very popular and the airframe data can be obtained
more easily, it was selected as an objective aircraft for the control law design.
Finally, a ¼ scale Piper J-3 cub model is selected and the aerodynamics
coefficients are calculated by DATCOM and AVL. Based on the forces and
moments acting on the aircraft, the trim equilibrium was calculated for getting
proper dynamics coefficients for the selected flight conditions. With the aircraft
aerodynamics coefficients, the aircraft dynamics characteristics and flying
qualities are also analyzed. The model studied in this thesis cannot answer
level one flying qualities in the longitudinal axis, which is required by MIL-F-
8785C. The stability augment system is designed to improve the flying qualities
of the longitudinal axis.
The work for autopilot design in this thesis includes five parts. First, the whole
flight profile is designed to automatically control aircraft from takeoff to landing.
Second, takeoff performance and guidance law is studied. Then, landing
performance and trajectory is also investigated. After that, the control law
design is decoupled into longitudinal axis and later-directional axis. Finally,
simulation is executed to check the performance for the auto-controller.
