In this paper, real-time system identification of an unmanned aerial vehicle (UAV) based on multiple neural net- works is presented. The UAV is a multi-input multi-output (MIMO) nonlinear system. Models for such MIMO system are expected to be adaptive to dynamic behaviour and robust to environmental variations. This task of accurate modelling has been achieved with a multi-network architecture. The multi-network with dynamic selection technique allows a combination of online and offline neural network models to be used in the architecture where the most suitable outputs are selected based on a given criterion. The neural network models are based on the autoregressive technique. The online network uses a novel training scheme with memory retention. Flight test validation results for online and offline models are presented. The multi-network dynamic selection technique has been validated on real-time hardware in the loop (HIL) simulation and the results show the superiority in performance compared to the individual models. Behaviour of the UAV is quantified mainly in terms of the translational velocities and the angular rates with respect to a fixed frame of reference. The data from the sensors is processed and recorded on an on-board computer. This platform with the sensors and on-board computer is flown remotely to collect the six DOF data from it. The data from the sensors and the actuators are used to develop UAV simulation models and thereby used in the design of flight control system. The accuracy of the designed controller depends largely on the quality of the data collected. Hence, the sensors are calibrated at different conditions to obtain superior quality of data. The flight control system (FCS) for the UAV performs tasks similar to those executed by a pilot for manned aircrafts. Robust control techniques, capable of adapting themselves to the changes in dynamics of the platform are necessary for the autonomous flight. Such controllers can be developed with the aid of suitable system identification (ID) techniques. This system ID based on flight data is also necessary for understanding the dynamics of the UAVs. Numerous system identification techniques have been proposed for the modelling of nonlinear systems. Fuzzy identification (9), state space identification (10), frequency domain analysis (11), artificial neural networks (12) are some of the prominent ones. The ability of the neural networks to learn makes them suitable for nonlinear applications.