Using Neural Networks to Model Hysteretic Kinematics in Tendon-Actuated Continuum Robots

TL;DR

This paper compares neural network models, including FNN, FNN with history, and LSTM, to accurately capture hysteretic and rate-dependent behaviors in tendon-actuated continuum robots.

Contribution

It evaluates and demonstrates the effectiveness of different neural network architectures in modeling hysteresis in continuum robots, highlighting the importance of input selection.

Findings

FNN with history buffer and LSTM effectively model hysteresis.

Model performance varies with robot design and input choices.

LSTM captures temporal dependencies comparable to history-augmented FNN.

Abstract

The ability to accurately model mechanical hysteretic behavior in tendon-actuated continuum robots using deep learning approaches is a growing area of interest. In this paper, we investigate the hysteretic response of two types of tendon-actuated continuum robots and, ultimately, compare three types of neural network modeling approaches with both forward and inverse kinematic mappings: feedforward neural network (FNN), FNN with a history input buffer, and long short-term memory (LSTM) network. We seek to determine which model best captures temporal dependent behavior. We find that, depending on the robot's design, choosing different kinematic inputs can alter whether hysteresis is exhibited by the system. Furthermore, we present the results of the model fittings, revealing that, in contrast to the standard FNN, both FNN with a history input buffer and the LSTM model exhibit the capacity to model historical dependence with comparable performance in capturing rate-dependent hysteresis.