Implementation of learning motion to control a robotic arm using haptic technology

Robotic arm motion learning is one of the most important recent robotics research area which brings robots in human life. This study is divided into three parts; a new approach for Human Robot Interaction (HRI) is proposed by assisting haptic technology hypothesis, a novel inverse kinematics solution for robotic arm is designed based on Artificial Neural Network (ANN), and motion control is implemented for Robotic Assisted Thoracoscopic Surgery (RATS) based on the kinematics of Remote Center of Motion (RCM). A new controller structure has been introduced by using virtual spring control method by compensating tool inertia effect. The applied force and torque are transformed to the desired position/orientation through the simultaneous matching between the human direct guidance and robot response. The novelty of the proposed ANN is that of including the current joint angles configuration as well as the desired position and orientation in the input pattern of ANN. The traditional ANN has got only the desired position and orientation of the end effector in the input pattern of ANN. The motion of surgical robot is constrained by the kinematics of RCM, so a new control design for RCM is introduced. The controller is implemented to satisfy the requirement of RATS. The control method is then verified. The comprehensive experimental results have shown significant improvement in learning performance and reducing motion errors. The inclusion of current joint angles configuration in ANN significantly increased the accuracy of estimation of the joint angles output. The results have proved the applicability and the efficiency of the proposed design in robotic surgery, especially in RATS.