Affiliations: Department of Electrical and Robotic Engineering, Shahrood University, Shahrood, Iran
Note: [] Corresponding author. Mohammad Mehdi Fateh, Department of Electrical and Robotic Engineering, Shahrood University, University Street, 3619995161 Shahrood, Iran. Tel./Fax: +98 23 32393116; E-mail: [email protected]
Abstract: A robot manipulator including its actuators is a complex system due to the nonlinearity, uncertainty and joint flexibility. To overcome complexity, this paper develops a novel indirect adaptive fuzzy control for electrically driven flexible-joint robot manipulators in which a novel estimation technique is introduced to estimate the uncertainty. The control structure differs from the previous ones due to using the voltage control strategy instead of the torque control strategy. The control design has only one control loop whereas the commonly used control design employs two control loops. Moreover, it is a decentralized control whereas the torque-based control design is a centralized control. The required feedbacks for each controller are the motor current, joint position, joint velocity and motor velocity. Some advantages are: considering the whole robotic system including robot and its actuators as a fifth-order model, simplicity of control design, computational efficiency, good performance, and guaranteed stability. Simulation results and comparisons show the effectiveness and superiority of the control approach over two control approaches. The first approach is a modified fuzzy proportional-integral control as a torque-based control method and the second one is a robust control as a voltage-based control method.
Keywords: Indirect adaptive fuzzy control, flexible-joint robot, voltage control strategy, actuator dynamics
