Research on position control of bionic waist joint driven by pneumatic artificial muscle

Abstract

As the most critical part of the overall structure of imitation human robots, bionic waist joint needs to be good and flexible in addition to the basic motion function of the waist joint to ensure that when it is applied to imitating robots, it can be more consistent with the physiological characteristics of the human waist joint. This paper proposes a bionic waist joint system composed of two bionic waist joint units based on the 3-DOF (degree of freedom) parallel mechanism driven by pneumatic artificial muscle. The kinematic analysis of the bionic waist joint unit mechanism and system is carried out, respectively. For the non-linear characteristics of pneumatic artificial muscle, its mathematics model is optimized. The dynamics model of bionic waist joint is derived. For the bionic waist joint’s position control, the PID controller and the RBF-PID controller are designed, and the proportional integral differential (PID) control algorithm and radial basis function neural network (RBF) tuning PID parameter control algorithm are studied, respectively. The simulation results show that both PID control and RBF-PID control achieve a relatively ideal effect on joint position control, but the BPPID control algorithm is better than PID control.