|
Abstract: |
This paper mainly presents a PM multi-DOF actuator for robot in-wheels drive applications and its control method. The torque model is established based on the finite-element model of the single pair poles in 3D finite element method software. Due to the special structure of the actuator, the Euler angles are adopted for deriving the kinematics and dynamic model. To reduce the effects of uncertainties of modeling error, nonlinear friction and external disturbances in the system, an approximation of neural network adaptive control method is applied to the actuator. The computation and simulation results show that the proposed analysis and control method can effectively derive the characteristics and improve the motion performance, which provides the primary theoretical guide for the configuration design, optimization and control research of multi-DOF deflection type actuators. |
Key words: deflection-type multi-DOF PM actuator neural network control |
DOI:10.11916/j.issn.1005-9113.2013.02.019 |
Clc Number:TM301 |
Fund: |