Shape Recognition of a Tensegrity With Soft Sensor Threads and Artificial Muscles Using a Recurrent Neural Network

Li, Wen-Yung; Takata, Atsushi; Nabae, Hiroyuki; Endo, Gen; Suzumori, Koichi

doi:10.1109/lra.2021.3091384

Cited by 13 publications

(12 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…is simple test showed the feasibility of people using computers to translate at that time. Since then, people began to study machine translation [3]. Abebe and others deepened the reform of Halliday system grammar and proposed the SCHRDLU model.…”

Section: Literature Reviewmentioning

confidence: 99%

Design of English Translation Model Based on Recurrent Neural Network

Wang

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

In order to improve the accuracy and stability of English translation, this paper proposes an English translation model based on recurrent neural network. Based on the end-to-end encoder-decoder architecture, a recursive neural network (RNN) English machine translation model is designed to promote machine autonomous learning features, transform the distributed corpus data into word vectors, and directly map the source language and target language through the recurrent neural network. Selecting semantic errors to construct the objective function during training can well balance the influence of each part of the semantics and fully consider the alignment information, providing a strong guidance for the training of deep recurrent neural networks. The experimental results show that the English translation model based on recurrent neural network has high effectiveness and stability. Compared with the baseline system, it has improved about 1.51–1.86 BLEU scores. Conclusion. The model improves the performance and quality of English machine translation model, and the translation effect is better.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Design of English Translation Model Based on Recurrent Neural Network

Wang

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…For this reason, we focused on the development of a very light soft tensegrity structure driven by thin McKibben muscles [13], which have a large power output relative to their own weight and flexible movement among other lightweight artificial muscles [14]. We have previously reported on the success of tensegrity structure shape recognition by incorporating a soft thread sensor and processing the data using recurrent neural network (RNN) [15]. This allows the robot to recognize the spatial shape of the environment based on the knowledge of its own shape as it moves through a space with its body in contact with a wall assuming that the robot moves through a space of the approximate size of the robot's motion range.…”

Section: Pneumatic Tubementioning

confidence: 99%

“…Such motion cannot actively generate horizontal or gravitational thrust, or passively adapt to the environment. Alternatively, robots that use artificial muscles to move tensegrity structures have also been studied [15][19] [20][21], but the amount of deformation that they can achieve is insufficient for the desired motions. Thus, it is necessary to develop a new method that allows a large amount of displacement to be generated in the tensegrity structure using artificial muscles.…”

Section: Pneumatic Tubementioning

confidence: 99%

“…2. Our group has previously demonstrated that the shape recognition of tensegrity can be achieved by applying sensor threads as the rubber threads of the tensegrity [15].…”

Section: A Proposed 4/3 Muscle Winding Conceptmentioning

confidence: 99%

“…According to [15], the shape of the tensegrity cannot be estimated if the rubber threads are slack. So the natural length of the rubber thread should be minimized.…”

Section: Experimental Evaluationsmentioning

confidence: 99%

See 2 more Smart Citations

Soft Tensegrity Robot Driven by Thin Artificial Muscles for the Exploration of Unknown Spatial Configurations

Kobayashi

Nabae

Endo

et al. 2022

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

The primary role of a robot exploring an unknown space is to investigate the state and the spatial shape of the environment. We have designed a soft robot that aims to move forward in an unknown space as it recognizes and adapts to the spatial shape of the environment. We previously reported that soft tensegrity and recurrent neural network can be used to realize tensegrity structure shape recognition. In this study, a tensegrity robot was designed to actively generate propulsive force as it presses its body against a wall in its surrounding environment. This robot design includes a novel artificial muscle arrangement called "4/3 muscle winding," which induces large deformation in the tensegrity structure. The application of this new artificial muscle arrangement allows two types of large deformations to be induced in the tensegrity structure, which results in displacements of 20% to 40% in the axial and radial directions. We have demonstrated that the robot, which was created by connecting the tensegrity structures, is lightweight and possesses passive shape adaptability in a three-dimensional environment. This tensegrity robot could enter an unknown space, such as a cave, and recognize the spatial shape of the surrounding environment by recognizing the tensegrity structure shape.

show abstract

Bioinspired Continuum Robots with Programmable Stiffness by Harnessing Phase Change Materials

Zhang¹,

Wang²,

Chen³

et al. 2023

Adv Materials Technologies

View full text Add to dashboard Cite

are referred to as continuum robots. [1] Compared with traditional robots composed of rigid links and joints, continuum robots featured with both softness and compliance exhibit reduced complexity in interactions with the environment in application scenarios, such as minimally invasive surgery, [2] environment exploration, [3] and safe human-robot interaction. [4] As a promising candidate for constructing safe human-centered robots, continuum robots are shifting the mechanical design of intelligent machines from the sole use of rigid materials toward the use of compliant materials. [5] However, these robots are challenged by the inherent contradiction between softness and load-bearing capacity when dealing with stress application scenarios, since the carrying capacity of compliant materials is much worse than that of rigid materials. [6] Aiming at this inherent challenge in existing continuum robots, a hybrid robotic system that combines both soft and rigid materials may figure out the contradiction between the structural stiffness and shape adaption capability. [7] Tensegrity structures, created by Fuller, may provide a novel paradigm for the continuum robotic design. [8] The combination of rigid components and high softness endows this structure with desirable characteristics found in both classically rigid and Continuum robots offer significant advantages over traditional ones in some specific scenarios, such as urban search and rescue, minimally invasive surgery, and inspection of cluttered environments. However, motions and/or operations of existing continuum robots always suffer from those limitations in varying curvature interaction scenarios because of the homogeneity and singleness of the structural stiffness. Herein, inspired by the mechanism of an elephant trunk for regulating local stiffness, a three-segment continuum robot constructed by tensegrity structure, which relies on a stiffness tunable material, with its Young's modulus switchable between 1.79 and 271.62 MPa to achieve the robotic stiffness programmable characteristics, is proposed. For predicting the robotic configuration with varying stiffness distribution, a mechanical model based on the framework of the finite element method is derived. Theoretical predictions reveal that the curvature of each segment can be regulated by programming stiffness of the smart materials; therefore, the customizable design can offer an effective route for real-time robotic interactions. By evaluating motion characteristics, stiffness performance, and conformal interaction capability, the experimental results demonstrate that the robot can freely regulate the configuration on-demand, which may provide a foundation for the application of continuum robots with programmable stiffness for interacting with unstructured environments.

show abstract

Shape Recognition of a Tensegrity With Soft Sensor Threads and Artificial Muscles Using a Recurrent Neural Network

Cited by 13 publications

References 27 publications

Design of English Translation Model Based on Recurrent Neural Network

Design of English Translation Model Based on Recurrent Neural Network

Soft Tensegrity Robot Driven by Thin Artificial Muscles for the Exploration of Unknown Spatial Configurations

Bioinspired Continuum Robots with Programmable Stiffness by Harnessing Phase Change Materials

Contact Info

Product

Resources

About