A Real-Time Landing Gear Simulation Model For A Very Light Aircraft: Development And Experimental Validation

The main purpose of the paper is to present a landing gear model for autonomous on ground control design and Hardware In The Loop Real-Time simulations. This model is a component of the real-time aircraft simulator, developed by CIRA (the Italian Aerospace Research Centre) for its fixed wing flying platform used to in-flight test all the developed UAV functionalities and, in particular, the autonomous take-off and post touch down algorithms. The autonomous take-off and post touch down managing systems are part of the autonomous GNC (Guide Navigation and Control) prototype worked out by CIRA, in the framework of the national founded project TECVOL (Technologies for the Autonomous Flight). The landing gear model core, implemented in MATLAB/Simulink environment, is a set of flexible physical-based equations that can easily simulate different aircrafts and runways by changing few design parameters. Moreover, the model contains even the logic for managing the dynamics of the vehicle in case of velocity near to zero. It is coupled with a six degree of freedom aircraft rigid body dynamics equations also implemented in Simulink. The landing gear model inputs are: runway elevation; aircraft inertial state variable (position, velocity and attitude); nose gear steering angle; brake command; forces and moments in body axis whereas the outputs are the gear forces and moments in body axis and a Boolean signal that simulates the Weight on Wheels (WoW) sensors. For each wheel of the landing gear is defined the position in body frame. This assignment, along with runway elevation and aircraft inertial state variable, allows to calculate the velocity, position, and contact point in the inertial frame besides forces and moments, due to the interaction between tires and ground. A relevant advance in the use of a model that calculates the ground reaction separately for each wheel is the capacity to simulate the unbalanced dynamic loads normally encountered during on-ground aircraft operations. The normal tire force is computed as typical spring and damper reaction depending on the distance and relative velocity between the wheel and ground. Axial and lateral friction forces are computed using an approximation of tire theory and model reported from Pazmany (1986). The model is able to evaluate friction forces and moments at all velocities varying the values of friction coefficients. Probably the most difficult ground dynamics phenomenon to simulate in real-time is stopping the aircraft [2]. When the aircraft is stopped the friction forces and moments must be equal to the other external forces. At very low velocity, close to zero, if the usual equations of tire dynamics are applied, it can be computed unrealistic forces and moments due to some divisions by speed present in the utilized formulae. Our proposed solution foresees that the friction forces and moments magnitude decrease towards the external forces and moments when the velocity is decayed to zero. After the vehicle stop, only when the external forces (aerodynamic, thrust and weight forces) exceed the static friction reaction, the aircraft starts to move. To validate the model and to estimate the friction coefficient we designed specific experimental test to perform with aircraft on the runway. The comparison of experimental and simulated data demonstrates that the proposed model is able to describe the real landing gear behaviour. The model has been used to design and verify the on-ground control law of UAV GNC system.