Abstract
This paper deals with the effectiveness of a neural network controller implemented in a control system of an underwater robot manipulator when the robot performs the task of catching and gasping an object floating in water. The neural network must learn the nonlinear behavior of the floating object generated by the fluid flow due to the motion of the robot arm when the robot rotates and extends its arm toward the object. The learnd neural network is implemented as a forward controller together with a feedback controller. Experiments showed a good performance in catching and grasping the floating object in water even when the object is moving away.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
