The present work studies the comparison of a modified reference controller based on an artificial neural network with a proportional-integral-derivative (PID) controller in the modeling and controlling of a two-degree-of-freedom (2DOF) robotic arm. In industrial automation and other domains, autonomous machines are important factors that require accurate control to adjust to their nonlinear dynamics. When dealing with this complexity, traditional controllers like proportional-integral-derivative (PID) controllers often deal with difficulties. The primary goal is to investigate. The primary aim is to investigate the performance of each controller. The process includes using Simulink to develop a closed-loop kinematic model of the 2DOF robotic arm. With the goal of anticipating the joint angles required for tracking desired trajectories, an artificial neural network (ANN) controller was built and trained using simulated data. For the purpose of determining the efficiency of the ANN-based controller against a conventional PID controller, simulations were run. The main metrics that were examined were tracking accuracy, reaction time, and stability. The closed-loop feedback mechanism provided the ANN with the capacity to continuously adapt its predictions and provide accurate results. The tracking accuracy of the ANN-based controller was found to be higher than that of the PID controller, with less error. Additionally, the ANN demonstrated to be more adaptive, preserving stability, while the PID controller failed. These demonstrated the ANNs' capability to adjust in real time to dynamic changes, offering a more accurate and responsive control system. In conclusion, specifically for complicated situations, the ANN-based control system proves to be a viable substitute for traditional PID controllers in robotic arm control. The findings suggest that ANNs may give increased performance in nonlinear and dynamic systems. Highlights 1. Controller comparison: For a two-degree-of-freedom robotic arm, the study described here performs a detailed comparison between a traditional PID controller and an artificial neural network-based Model Reference Controller. 2. Adaptable control system: This research has shown that, in situations with complicated nonlinear dynamics and in the presence of external disruptions, control based on artificial neural networks (ANNs) is more responsive and durable than traditional control carried out by PID controllers. 3. Performance metrics: Since the ANN-based model reference control (MRC) outperforms the PID controller in terms of accuracy and error reduction, tracking accuracy, reaction time, and reliability are the most important performance metrics. 4. Implications for advanced robotics: The perspectives into future progress in robotics automation arise from the likelihood that ANN-based controllers are going to surpass traditional PID controllers in applications connected to very complex, real-time, and dynamic environments.