이력 현상 크기에 따른 텐던-쉬스 메커니즘의 RNN 기반 이력 현상 모델

Abstract

The tendon-sheath mechanism is one of the key components of flexible surgical robots. It is essential for enabling remote actuation systems within the design of flexible surgical robots, thereby addressing and overcoming constraints related to size limitations. However, it simultaneously introduces a critical issue, as it causes hysteresis, which impedes the precise movements required in surgical robotics. Previous research has successfully predicted the magnitude of hysteresis based on the path of the tendon-sheath mechanism. In contrast to earlier studies, the objective of this research is to predict the behavior of hysteresis using hysteresis magnitude data. To build the hysteresis prediction model, a combination of a Recurrent Neural Network (RNN) and the Preisach model was employed. During the training process, both the hysteresis magnitude and the desired displacements are used to develop a model capable of predicting the actual output displacements. The proposed model estimated hysteresis with an RMSE within 0.0441 mm. This approach is expected to improve the reliability and performance of flexible surgical robots.