SELF-LEARNING OF DELTA ROBOT USING INVERSE KINEMATICS AND ARTIFICIAL NEURAL NETWORKS

As known as Parallel-Link Robot, Delta Robot is a kind of Manipulator that consists three arms mounted in parallel. has central joint constructed an end-effector represented gripper. An Analysis Inverse Kinematic (IK) used to convert the trajectory (X, Y) into rotations stepper motors (ZA, ZB and ZC). The proposed method Artificial Neural Networks (ANNs) simplify process IK solver. solver generated datasets contain motion data robot. There are 11 KB Datasets consist 200 be trained. was trained 58.78 seconds 5000 iterations. Using learning rate (α) 0.05 produced average accuracy 97.48%, loss 0.43%. also tested transfer over Socket.IO with 115.58B 6.68ms.