Comparative Study between Robust Control of Robotic Manipulators by Static and Dynamic Neural Networks

Table 2

Advantages and limitations of the neural network control approaches.

Control strategy	Advantages	Limitations
NN controller for improvement of a classic controller PD	(i) Initialization of the NN weights is arbitrary	No reliable response of the system, seen in Figure 7, due to the peak in the torques response, facesdisturbances and robotic uncertainties
	(ii) Online learning of the NN
	(iii) No offline training requirements
	(iv) Approximation, by the neural network, of the function which gathers the nonlinearities and uncertainties included in the robot dynamics
	(v) Overcoming some limitation of the conventional controller PD
	(vi) Guaranteed stability in presence of nonlinearities and uncertainties
NN controller via a high order dynamic neural network	(i) An exact online identification of the robot state thanks to the dynamic neural network	(i) Initialization of the NN weights is not arbitrary; (ii) Offline training requirements to find the suitable initial NN weights values
	(ii) Guaranteed stability in presence of nonlinearities and uncertainties
	(iii) Reliable response of the system faces disturbances and robotic uncertainties (Figures 12 and 13)