A novel dynamic decoupling control algorithm for PMSA by utilizing single neural adaptive controllers

This paper proposes a permanent magnet spherical actuator (PMSA) with three-dimensional magnet array. The application of 3D magnet array improves the output torque significantly. However, as PMSA is a multivariable and nonlinear system, it is very complicate to achieve high accuracy control. Therefore, a novel single neural adaptive control algorithm is proposed in this paper. Compared with the PD and PID control methods, the proposed algorithm is able to realize self-adaptive to the perturbation, which improves the motion stability of PMSA a lot. Euler angles are applied to describe the orientation of rotor. Lagrange equations are utilized to establish the dynamic model of the actuator. As the center of gravity does not coincide with the center of rotor sphere, gravity torque is taken into consideration to enhance the precision of dynamic model. The stability of this algorithm is validated with detailed derivation. Simulation model is formulated in Matlab/Simulink. It proves that the proposed algorithm can make the system track the continuous trajectories in high accuracy.