Fernando Passold

This paper describes experimental results applying artificial neural networks to perform the position control of a real scara manipulator robot. The general control strategy consists of a neural controller that operates in parallel with a conventional controller based on the feedback error learning architecture. The main advantage of this architecture is that it does not require any modification of the previous conventional controller algorithm. MLP and RBF neural networks trained on-line have been used, without requiring any previous knowledge about the system to be controlled. These approach has performed very successfully, with better results obtained with the RBF networks when compared to PID and sliding mode positional controllers.

This paper describes experimental results applying artificial neural networks to perform the position control of a real scara manipulator robot. The general control strategy consists of a neural controller that operates in parallel with a conventional controller based on the feedback error learning architecture. The main advantage of this architecture is that it does not require any modification of the previous conventional controller algorithm. MLP and RBF neural networks trained online have been used, without requiring any previous knowledge about the system to be controlled. The approach has performed very successfully, with better results obtained with the RBF networks when compared to PID and sliding mode positional controllers.

This paper describes experimental results achieved by applying artificial neural networks (NNs) in trying to perform force/position control of a real Scara manipulator robot. The general control strategy consists of a neural controller that operates in parallel with a conventional controller based on feedback error learning architecture. Two NNs are used: one for the position loop control and another to the force loop control based on the structure known as the hybrid force/position control. The main purpose was to let the NNs compensate the dynamical effects that arises when a manipulator is in contact with an environment. Successful results have been achieved for the position loop control but practical problems were observed which are related to the force loop control using NNs even with different sets of input data for this NN. Practical problems are discussed.