During recent years, interest in Unmanned Aerial Systems (UAS) has shown an enormous growth in both military and civil aviation. The increased demand has led engineers and designers to search for methods to improve flight performance, especially for long endurance reconnaissance and intelligence missions. However, the validation of a performance improvement technique requires a high number of flight tests, which can be very demanding in terms of both time and money. A high-level simulation model provides an alternative solution, allowing engineers to perform numerical calculations to test new aircraft designs or any modifications to existing ones in a simulation environment.
Designing a model or realizing an aircraft simulator may, under certain conditions, result in aberrant results, including numerical instability due to the successive error increases. To cope with this difficulty, the aircraft model is divided into sub-models. The general model of the aircraft depends on its geometry, its systems and the environmental factors. Therefore, its overall architecture is composed of aerodynamics, propulsion and actuation systems, as well as its mass and inertia. Thus, the modelling procedure for an aircraft is a collection of methods for estimation of each sub-model. Several studies have been conducted to examine this methodology.
Jodeh et al.developed a nonlinear simulation model to estimate the flight dynamics of the Rascal, with its aerody- namic model designed using the DATCOM procedure. The propeller model was based on the airfoil characteristics while the engine model consisted of a linear lookup table. The mass and inertia analyses were conducted by the experimental pen- dulum method. Al-Radaideh designed and built a test bed for the ARF60 AUS-UAV. The model was constructed under Simulink using Aerosim and Aeroblockset to facilitate the flight control system development. The aerodynamics was modelled using linear estimation based on the aircraft’s geometry. The propulsion model used a transfer function with the throttle command as input, and the RPM of the engine and the thrust produced by the propeller as outputs. This model was used to test autopilot behaviour. The results have shown that the outputs were very close to the command values.
A procedure to model small unmanned vehicles at high angles of attack was presented by Selig. This methodology was developed for UAV/Radio-controlled Aircraft (RC). The UAV/RC was divided into basic components, such as wing, horizontal tail and vertical tail, in order to evaluate their inter- action effects. The aerodynamic analysis was performed using strip theory while the propeller model was estimated from blade element momentum theory using PROPID code.The aircraft model was implemented in the Flight Simulator (FS- 1) to determine its flight dynamics at stall condition.
Elharouny et al. provided a procedure for modelling small UAV. This procedure was applied on a Sky Raider Mach . The aerodynamic modelling was performed by coupling Xfoil to determine the airfoil aerodynamics characteristic and DATCOM to estimate the overall aerodynamic model of the UAV.The propulsion model consisted of evaluating the thrust performance of the UAV. It was estimated experimentally using a spring scale to measure the thrust force along with a set of throttle command and incoming wind speed. The moment of inertia and the center of gravity were obtained from a pendu- lum method while the mass were determined using a balance. The resulting model was used for control design tasks.
Kamal et al.presented a flight simulation model for a small commercial off-the-shelf UAV/RC, the ‘‘tiger Trainer”. The structural model consisted in determining the mass, the center of gravity and the moment of inertia of the UAV. The mass was obtained using an accurate digital scale and the center of gravity was estimated from a moment balance about the nose wheel. The pendulum method was thus applied to experimentally evaluate the UAV moment of inertia. The propulsion system consisted of a piston-propeller engine. The propulsion modelling was separated into the propeller analysis and the engine dynamic estimation. The propeller analysis was performed experimentally in a low speed wind tunnel to mea- sure thrust and power performance from static condition to windmill regime. The engine dynamic was built from a black box using pulse on the throttle as input and engine rotation speed as output. The aerodynamic characteristics were obtained, in the first step, by analysing the similarity of the wing airfoil with conventional airfoil as Clark-Y. In the second step, DATCOM was used to obtain aerodynamic behaviour of the entire UAV. The actuator was modelled from an identification process on a servomotor. This methodology required a time history of the rotational angle of the servomotor as function of a signal inputs which were measured experimentally. The complete six DoF nonlinear model of the UAV was assembled using MATLAB/Simulink. The model was verified, for a horizontal steady flight, on its longitudinal and lateral dynamic. The results showed a good agreement with the experimental flight test...