Neural network modelling of robots is introduced using the GL matrices
and operator (Ge et al. 1994), and a new adaptive neural network cont
roller for robots is presented. The controller is based on direct adap
tive techniques, and there is no need for matrix inversion. Unlike man
y neural network controllers in the literature, inverse dynamical mode
l evaluation is not required and no time-consuming training process is
necessary, except for initializing the neural networks based on appro
ximate parameters of the initial posture at time t = 0. It is shown th
at if gaussian radial basis function networks are used, uniformly stab
le adaptation is assured and asymptotic tracking is achieved.