Reliability Evaluation Method for Parallel Mechanisms Kinematics under Time-Varying Loads

Under the action of time-varying dynamic loads, the deformation caused by the degradation of the linkage stiffness in parallel mechanisms and the deformation caused by impact responses are coupled, resulting in high uncertainty that reduces motion stability and accuracy. This paper proposes a motion reliability evaluation method that considers the propagation of epistemic uncertainty. The uncertain Liu process is employed to model the uncertain degradation process of key connecting components in parallel mechanisms. The inverse uncertainty distribution is utilized in both forward and inverse kinematics models to calculate the impact of structural degradation on kinematics reliability. In the case study involving a humanoid robot's ankle joint, the proposed reliability evaluation method is compared with a probability-based approach. The results indicate that the proposed method would provide a more conservative maintenance strategy.