Mixed model-based/neural network H∞ impedance control of constrained manipulators

This paper deals with impedance H_∞ control problem of constrained manipulators based on neural network techniques. With this approach, it is guaranteed that the relationship between the force and position errors converges to a given dynamic behavior. The neural networks proposed in this paper adapt only the uncertain dynamics of the robotic manipulator, they actuate as complement of the nominal model. The disturbance rejection problem, formulated through the H_∞ performance index, comprehend the position errors and the interaction forces between the manipulator end-effector and the environment. Simulated results obtained from a planar robot manipulator under a constrained movement are presented.