Development of an attitude control system of a heavy-lift hexacopter using Elman recurrent neural networks

Abstract

Hexacopter is a type of multicopter that can be used to lift a heavy load, hence very convenient to be utilised in agricultural fields. As the consequence, however, the attitude control of this hexacopter is rather difficult compare with that of a quadcopter with four motors, due to gyroscopic effect of the additional motors and in its combination with the heavy loads. In this paper, we have developed a direct inverse controller system using an Elman neural networks for the attitude and altitude control of the hexacopter. Experiments are conducted using a flight data taken from a test-bed system. Results show that the attitude characteritics of the heavy-lift hexacopter can be controlled successfully, especially when an optimized Elman neural networks as the direct inverse controller system is utilized.