Neural Network and Jacobian Method for Solving the Inverse Statics of a Cable-Driven Soft Arm With Nonconstant Curvature

Abstract

The solution of the inverse kinematics problem of soft manipulators is essential to generate paths in the task space. The inverse kinematics problem of constant curvature or piecewise constant curvature manipulators has already been solved by using different methods, which include closed-form analytical approaches and iterative methods based on the Jacobian method. On the other hand, the inverse kinematics problem of nonconstant curvature manipulators remains unsolved. This study represents one of the first attempts in this direction. It presents both a modelbased method and a supervised learning method to solve the inverse statics of nonconstant curvature soft manipulators. In particular, a Jacobian-based method and a feedforward neural network are chosen and tested experimentally. A comparative analysis has been conducted in terms of accuracy and computational time.

Comparison and experimental validation of predictive models for soft, fiber-reinforced actuators

“…Data-driven strategies have been used on physically realized systems to characterize dynamic behavior, and, in some cases, to develop control policies. Diverse schemes have been proven on a variety of soft systems, including neural networks (Gillespie et al, 2018;Giorelli et al, 2013Giorelli et al, , 2015Thuruthel et al, 2018), genetic algorithms (Giorgio-Serchi et al, 2017), and other regression techniques (Bruder et al, 2018b;Elgeneidy et al, 2018;Veale et al, 2018). Within first principles-based models, two subcategories emerge.…”

Learning Global Inverse Kinematics Solutions for a Continuum Robot

“…Moreover, the same approach can be used for learning the inverse statics of continuum robots. Researchers have been able to learn the inverse statics of continuum robots for a non-redundant case (Giorelli et al 2015). We believe that our approach can be easily extended for learning the inverse statics of a redundant continuum robot also.…”

Robust control of a silicone soft robot using neural networks

“…Indeed, it is not new to use neural network to approximate the model of soft robots. In [14], the researchers used neural network to solve the inverse statics of a cable-driven soft arm with non-constant curvature. [15] used FEM to train a neural network to get the model of modular soft robots.…”

“…Theorem 1. For the studied soft robot described by (12), if the neural network approximates with an accuracy index satisfying the following inequality < (14) where and are the characteristic parameters of soft robot defined in (4) and (7), then the proposed controller (13) can exponentially drive any ∈  to the desired constant position ∈ , i.e., lim →∞ ‖ ( ) − ‖ 2 = 0. Proof 1.…”