Terminal Sliding Mode Control of an Anthropomorphic Manipulator with Friction Based Observer

Abstract

The role of modern control techniques has been instrumental in today's robotic applications because of their increasing requirements for reliability, accuracy, productivity and repeatability. Robotic manipulators are highly non-linear systems with coupled dynamics and thus are vulnerable to a lot of disturbances such as unknown payloads, dynamics that the model and joint friction do not predict. To achieve superior performance and reliability in the presence of friction, this research proposes a robust control algorithm for a five-degree-of-freedom (DoF) robotic manipulator. The dynamic LuGre friction model is used to develop the robot's dynamic model. A sliding mode observer (SMO) is proposed for the estimation of the internal friction state of the LuGre model. The friction and load torque are based on an estimated state to compensate for unidentified friction. A Lyapunov candidate function is used to check the stability of the controller with the SMO. The proposed control methodology has been designed and implemented in MATLAB/Simulink environment to illustrate the tracking of various trajectories. This study's outcomes proved that the proposed control law with model-based friction compensation is effective and efficient.