Akademik Veri Yönetim Sistemi
BULLETIN OF THE POLISH ACADEMY OF SCIENCES. TECHNICAL SCIENCES, cilt.74, sa.1, ss.1-16, 2026 (SCI-Expanded, Scopus)
Recent advances in decision-making algorithms used in mobile robotics require more advanced and adaptive control strategies. Model predictive control (MPC) is one of the prominent strategies to manage diverse kinds of complex dynamic systems. Despite their widespread adoption in industrial robotics owing to their structural simplicity and ease of implementation, proportional-integral-derivative (PID) controllers exhibit notable limitations in effectively addressing process variations and system constraints, particularly those arising from mechanical constraints on joint positions and velocities. As autonomous mobile robots (AMRs) have been increasingly deployed in various and demanding applications, the need for more advanced control algorithms has become critical. In this study, a novel hybrid control framework integrating MPC and PID strategies is proposed and experimentally validated on a real-world differential drive robot, aiming to enhance tracking accuracy and overall operational performance. The system model of the TurtleBot3 robot is identified using the System Identification Toolbox and validated through extensive motion tests on the real robot by using Robot Operating System 2. The proposed control scheme combines the predictive capabilities of MPC with the reactive nature of PID to facilitate improved management of system constraints, aiming to improve the performance of AMR in controlling both linear and angular velocities. Experimental results show that the hybrid MPC-PID controller exhibits better performance by reducing tracking errors while maintaining reliability and robustness characteristics over a conventional PID controller. These results demonstrate that the hybrid MPC-PID approach provides a more effective solution for dynamic control tasks in mobile robotic systems, particularly in scenarios requiring high accuracy and reliability.