PD Control of Robot with RBF Networks Compensation

In this paper the popular PD controller of robot manipulator is modified. RBF neural networks are used to compensate the gravity and fi-iction. No exact knowledge of the robot dynamics is required. The euggeated learning law of neuro compensator is similar to the well-known backpropagation algorithm but wit h addit ional robust terms. Lyapuuov-liie analysis is used to derive the stability of learning algorithm.

• Type: Conference
• Year: 2000
• Where: IJCNN
• Authors: Wen Yu, José Antonio Heredia
• Keywords: IJCNN 2000, Neural Networks, Popular Pd Controller, RBF Neural Networks, Well-known Backpropagation Algorithm