In the paper, in order to deal with the attitude control problem of a rigid body in a 3-D space, a new control strategy in hypercomplex algebra is developed. The proposed approach is based on two parallel controllers derived in quaternion algebra. The first one is a feedback controller of PD type, while the second is a feed-forward controller implemented by means of an hypercomplex multilayer perception (UMLP) neural network. Quaternion algebra allows to simplify the computational complexity of the controllers and leads to a more efficient learning algorithm for the neural network. Several simulations and comparisons with other control strategies show the suitability of the proposed approach.
Attitude feedforward neural controller in quaternion algebra
XIBILIA, Maria Gabriella
1999-01-01
Abstract
In the paper, in order to deal with the attitude control problem of a rigid body in a 3-D space, a new control strategy in hypercomplex algebra is developed. The proposed approach is based on two parallel controllers derived in quaternion algebra. The first one is a feedback controller of PD type, while the second is a feed-forward controller implemented by means of an hypercomplex multilayer perception (UMLP) neural network. Quaternion algebra allows to simplify the computational complexity of the controllers and leads to a more efficient learning algorithm for the neural network. Several simulations and comparisons with other control strategies show the suitability of the proposed approach.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
