Efficiency assessment of SOA-based computed torque control: A comparative analysis with NE-based approach

Abstract

This paper proposes the spatial operator algebra-based computed torque control scheme applied to an anthropomorphic 3-degree of freedom robotic manipulator, which aims to reduce the computational cost of the classical methods and integrate the advantages of low computational cost into advanced robotic control systems. The computational cost is increased due to the calculations of the inverse dynamic and the feedback control loop. The spatial operator algebra (SOA) algorithm provides a systematic derivation, evaluation, and subsequent conceptual interpretation of the manipulator dynamics model. This paper presents a powerful and efficient solution for controlling the dynamics and trajectory of the manipulator. In order to show the efficiency of the solution, the Newton Euler (NE) based control schemes are also applied to the manipulator. The SOA-based controller significantly reduced the computational cost and performed approximately 60%-70% faster than the NE-based controller. Furthermore, different initial states, disturbances, and uncertainty tests are implemented and the SOA-based controller demonstrated successful performance under various conditions while maintaining a lower computational cost. In this study, the advantages and limitations of each method (SOA-based, NE-based) are evaluated and the potential benefits of using SOA in the computed torque control scheme are highlighted. The SOA-based controller, which was verified by simulation, is then implemented in real-time and showed successful performance.