Filters








3,992 Hits in 7.5 sec

Research on the inverse kinematics prediction of a soft biomimetic actuator via BP neural network [article]

Huichen Ma, Junjie Zhou, Jian Zhang, Lingyu Zhang
2022 arXiv   pre-print
To overcome these intrinsic problems, we propose a back-propagation (BP) neural network learning the inverse kinetics of the soft biomimetic actuator moving in three-dimensional space.  ...  to realize the inverse kinematic control of the soft biomimetic actuator.  ...  These models may not accurately capture the complex nonlinear dynamics of soft robots.  ... 
arXiv:2110.13418v2 fatcat:ybvovx6kfzet5hnsiyxoo234om

High-bandwidth nonlinear control for soft actuators with recursive network models [article]

Sarah Aguasvivas Manzano, Patricia Xu, Khoi Ly, Robert Shepherd, Nikolaus Correll
2021 arXiv   pre-print
This technique allows for reduced model sizes and increased control loop frequencies when compared with conventional RNN models.  ...  We present a high-bandwidth, lightweight, and nonlinear output tracking technique for soft actuators that combines parsimonious recursive layers for forward output predictions and online optimization using  ...  Killpack, “Learning nonlinear dynamic models of soft robots for model predictive control with neural networks”, 2018 IEEE International Conference on Soft Robotics (RoboSoft), Livorno, 2018,  ... 
arXiv:2101.01139v1 fatcat:3dlvxlnjivb5lii3cfwlc7pqa4

Smith Predictor Type Control Architectures for Time Delayed Teleoperation

Andrew C. Smith, Keyvan Hashtrudi-Zaad
2006 The international journal of robotics research  
The proposed architectures consist of a novel pseudo two-channel nonlinear predictive controller and its variations that use neural networks for online estimation of the slave and environment dynamics  ...  An early control methodology for time delayed plants is the Smith predictor, in which the plant model is utilized to predict the nondelayed output of the plant and move the delay out of the control loop  ...  Acknowledgments This work was supported in part by Natural Sciences and Engineering Research Council of Canada (NSERC), Canada Foundation for Innovation (CFI) and Ontario Innovation Trust (OIT).  ... 
doi:10.1177/0278364906068393 fatcat:z6a3hp3j3bdfnafnzlxwl2naby

Nonlinear Systems Identification Using Deep Dynamic Neural Networks [article]

Olalekan Ogunmolu, Xuejun Gu, Steve Jiang, Nicholas Gans
2016 arXiv   pre-print
This paper investigates the effectiveness of deep neural networks in the modeling of dynamical systems with complex behavior.  ...  Recently, deep neural networks have proven to be very effective in pattern recognition, classification tasks and human-level control to model highly nonlinear realworld systems.  ...  as in § VI-D with the soft-robot and glassfurnace network.  ... 
arXiv:1610.01439v1 fatcat:r233vl4fajcnfmx5vettogmdsm

Neural Model Extraction for Model-Based Control of a Neural Network Forward Model

Shuhei Ikemoto, Kazuma Takahara, Taiki Kumi, Koh Hosoda
2021 SN Computer Science  
In this research, we propose a new approach-i.e., neural model extraction, that enables model-based control for a feed-forward neural network trained for a nonlinear state equation.  ...  Thus far, because neural networks are a radically different approach to mathematical modeling, control theory has not been applied to them, even if they approximate the nonlinear state equation of a control  ...  We assume that the NN was trained to learn the nonlinear state equation of a robot whose dynamics are extremely difficult to mathematically model, such as for soft and bio-inspired robots.  ... 
doi:10.1007/s42979-021-00456-4 fatcat:x36obdlovvftdgrnn2bywmall4

LEARNING SOFT ROBOT AND SOFT ACTUATOR DYNAMICS USING DEEP NEURAL NETWORK

Hari Prakash Thanabalan
2021 International Journal of Engineering Applied Sciences and Technology  
In this paper, modelling of the soft robot using finite element modelling will be discussed in conjunction with deep neural network for the bending and control of the end effector.  ...  Inspired by living organisms and being the forefront of robotics evolution, the research in soft robotics has been growing exponentially.  ...  However, comparing with a rigid robot, modelling a soft robot can be challenging due to their complex dynamics and nonlinearity.  ... 
doi:10.33564/ijeast.2021.v05i09.003 fatcat:knpeztdk55hm7jywsbh5urzagy

Industrial applications of soft computing: a review

Y. Dote, S.J. Ovaska
2001 Proceedings of the IEEE  
Fuzzy logic (FL), neural networks (NN), and evolutionary computation (EC) are the core methodologies of soft computing.  ...  Soft computing (SC) is an evolving collection of methodologies, which aims to exploit tolerance for imprecision, uncertainty, and partial truth to achieve robustness, tractability, and low cost.  ...  Chemical Process Industry McAvoy surveyed various kinds of neural networks applied to chemical processes for diagnosis, modeling, feedforward control, soft sensing, and nonlinear model prediction control  ... 
doi:10.1109/5.949483 fatcat:erzs7fqa35dixgm7mq2efptpai

Model-Based Control of Soft Actuators Using Learned Non-linear Discrete-Time Models

Phillip Hyatt, David Wingate, Marc D. Killpack
2019 Frontiers in Robotics and AI  
Deep neural networks are a powerful tool for modeling systems with complex dynamics such as the pneumatic, continuum joint, six degree-of-freedom robot shown in this paper.  ...  Overall, our results show the potential of combining empirical modeling approaches with model-based control for soft robots and soft actuators.  ...  Our specific contributions include the following: • Development of a non-linear neural network (NN) architecture for dynamic modeling of a 6 DoF pneumatic robot with soft actuators based on data from a  ... 
doi:10.3389/frobt.2019.00022 pmid:33501038 pmcid:PMC7805923 fatcat:unfm5zq4jzd6dj2fzqs3b2sjoa

Autonomous Tissue Manipulation via Surgical Robot Using Learning Based Model Predictive Control [article]

Changyeob Shin, Peter Walker Ferguson, Sahba Aghajani Pedram, Ji Ma, Erik P. Dutson, Jacob Rosen
2019 arXiv   pre-print
Two AI learning based model predictive control algorithms using vision strategies are proposed and studied: (1) reinforcement learning and (2) learning from demonstration.  ...  The complex dynamics of soft tissue as an unstructured environment is one of the main challenges in any attempt to automate the manipulation of it via a surgical robotic system.  ...  As opposed to the reviewed model-free approaches using linearization, the proposed algorithms directly learn nonlinear dynamics of tissue in image space and control the robot with nonlinear optimization  ... 
arXiv:1902.01459v2 fatcat:ii4wxoyc2fcilkxl6d4b34dyfq

Review of machine learning methods in soft robotics

Daekyum Kim, Sang-Hun Kim, Taekyoung Kim, Brian Byunghyun Kang, Minhyuk Lee, Wookeun Park, Subyeong Ku, DongWook Kim, Junghan Kwon, Hochang Lee, Joonbum Bae, Yong-Lae Park (+2 others)
2021 PLoS ONE  
followed by a summary of the existing machine learning methods for soft robots.  ...  However, compared to rigid robots, soft robots have issues in modeling, calibration, and control in that the innate characteristics of the soft materials can cause complex behaviors due to non-linearity  ...  Hyatt et al. proposed a predictive model based on the neural networks, and a learning method for the linearized discrete state space representation of soft robots [87, 88] . G.  ... 
doi:10.1371/journal.pone.0246102 pmid:33600496 pmcid:PMC7891779 fatcat:alu4zm72irespj6wydikzjb6ie

Neural Networks Predicting Microbial Fuel Cells Output for Soft Robotics Applications

Michail-Antisthenis Tsompanas, Jiseon You, Hemma Philamore, Jonathan Rossiter, Ioannis Ieropoulos
2021 Frontiers in Robotics and AI  
The use of neural networks, namely a nonlinear autoregressive network with exogenous inputs was employed to predict the electrical output of an MFC, given its previous outputs and feeding volumes.  ...  The development of biodegradable soft robotics requires an appropriate eco-friendly source of energy.  ...  NONLINEAR AUTOREGRESSIVE NETWORK WITH EXOGENOUS INPUTS Some applications of NARX networks are predictors, nonlinear filters or models of nonlinear dynamic systems.  ... 
doi:10.3389/frobt.2021.633414 pmid:33748191 pmcid:PMC7969642 fatcat:ch3a6trvy5acrijrghd7sktnwa

SOFT ROBOT POSITIONING USING ARTIFICIAL NEURAL NETWORK

Marko Kovandžić, Vlastimir Nikolić, Miloš Simonović, Ivan Ćirić, Abdulathim Al-Noori
2019 Facta Universitatis Series Automatic Control and Robotics  
The experiment investigated the performance of an artificial neural network in solving the inverse kinematic problem of a soft robot.  ...  The network was trained and tested using records collected at 200 randomly chosen robot positions. The relative testing error of positioning, about 5%, confirmed a predictable robot behavior.  ...  The purpose of the experiment was to prove that an artificial neural network can be trained to perform a predictable soft robot control by approximating the inverse kinematic problem.  ... 
doi:10.22190/fuacr1901019k fatcat:yq5cz63np5b7rdesvxn7zspima

Table of Contents

2022 IEEE/ASME transactions on mechatronics  
Tavakoli 645 Adaptive Neural Network Vibration Control for an Output-Tension-Constrained Axially Moving Belt System With Input Nonlinearity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ...  Xu 800 Neural-Dynamics Optimization and Repetitive Learning Control for Robotic Leg Prostheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ... 
doi:10.1109/tmech.2022.3162010 fatcat:eocxijxyovctpdf5iujxwgwpny

Soft robot perception using embedded soft sensors and recurrent neural networks

Thomas George Thuruthel, Benjamin Shih, Cecilia Laschi, Michael Thomas Tolley
2019 Science Robotics  
This approach enables the development of force and deformation models for soft robotic systems, which can be useful for a variety of applications, including human-robot interaction, soft orthotics, and  ...  However, both the soft sensors—and the encasing dynamical system—often exhibit nonlinear time-variant behavior, which makes them difficult to model.  ...  Other data for this study can be found in the database (https://github.com/mlsensor/SciRo).  ... 
doi:10.1126/scirobotics.aav1488 pmid:33137762 fatcat:xr2se6p6t5dp7mro2c6trxc3da

Editorial: Advances in Modeling and Control of Soft Robots

Concepción Alicia Monje Micharet, Cecilia Laschi
2021 Frontiers in Robotics and AI  
They address dynamic modeling and related control of soft robots, starting from the current approaches based on statics, or second-order dynamics, and model predictive control (MPC), using basic lumped-parameters  ...  The work by Hyatt et al. demonstrates that online model adaptation is key in soft robot dynamic modeling and shows their results with a model predictive control. It is based on Edited and reviewed by:  ...  ACKNOWLEDGMENTS The authors wish to acknowledge the contribution of speakers and attendees of the IEEE/RSJ IROS 2019 Workshop on "Advances in Modeling and Control of Soft Robots".  ... 
doi:10.3389/frobt.2021.706514 pmid:34095241 pmcid:PMC8176095 fatcat:xavqtjeulfggzebmia5tf6jso4
« Previous Showing results 1 — 15 out of 3,992 results