Generation of Inverse Kinematics(IK) Trajectory of 3 Link Manipulator using Deep Learning Model

Inverse kinematics (IK) is an inherent problem of manipulator robotic control, particularly for high degree-of-freedom (DoF) systems where traditional analytical and numerical approaches suffer from computational inefficiencies and instability. Tracking of any given trajectory requires follows the sequential data. This paper uses models based on recurrent neural networks (RNN), including RNN, long-short-term memory (LSTM), and bidirectional LSTM (BiLSTM), to solve the IK problem for a 3-link robotic manipulator. By utilizing the temporal dependencies inherent in trajectory data, these models effectively learn the mapping between end-effector positions and joint angles. Experimental results demonstrate that BiLSTM outperforms other models in terms of accuracy and generalization across different datasets, making it a promising solution for real-time trajectory tracking. The findings highlight the potential of deep learning based sequential models in improving the efficiency and adaptability of robotic motion planning.