A Fuzzy Deep Neural Network with Sparse Autoencoder for Emotional Intention Understanding in Human-Robot Interaction

2020 ◽  
pp. 1-1
Author(s):  
Luefeng Chen ◽  
Wanjuan Su ◽  
Min Wu ◽  
Witold Pedrycz ◽  
Kaoru Hirota
Keyword(s):  
Neural Network ◽  
Deep Neural Network ◽  
Human Robot Interaction ◽  
Robot Interaction ◽  
Intention Understanding ◽  
Sparse Autoencoder

scholarly journals A Bayesian Deep Neural Network for Safe Visual Servoing in Human–Robot Interaction

2021 ◽  
Vol 8 ◽  
Author(s):  
Lei Shi ◽  
Cosmin Copot ◽  
Steve Vanlanduit
Keyword(s):  
Neural Network ◽  
Visual Servoing ◽  
Deep Neural Network ◽  
Human Robot Interaction ◽  
Training Data ◽  
Robot Interaction ◽  
Field Methods ◽  
Distance Methods ◽  
Unseen Data ◽  
Interval Coverage

Safety is an important issue in human–robot interaction (HRI) applications. Various research works have focused on different levels of safety in HRI. If a human/obstacle is detected, a repulsive action can be taken to avoid the collision. Common repulsive actions include distance methods, potential field methods, and safety field methods. Approaches based on machine learning are less explored regarding the selection of the repulsive action. Few research works focus on the uncertainty of the data-based approaches and consider the efficiency of the executing task during collision avoidance. In this study, we describe a system that can avoid collision with human hands while the robot is executing an image-based visual servoing (IBVS) task. We use Monte Carlo dropout (MC dropout) to transform a deep neural network (DNN) to a Bayesian DNN, and learn the repulsive position for hand avoidance. The Bayesian DNN allows IBVS to converge faster than the opposite repulsive pose. Furthermore, it allows the robot to avoid undesired poses that the DNN cannot avoid. The experimental results show that Bayesian DNN has adequate accuracy and can generalize well on unseen data. The predictive interval coverage probability (PICP) of the predictions along x, y, and z directions are 0.84, 0.94, and 0.95, respectively. In the space which is unseen in the training data, the Bayesian DNN is also more robust than a DNN. We further implement the system on a UR10 robot, and test the robustness of the Bayesian DNN and the IBVS convergence speed. Results show that the Bayesian DNN can avoid the poses out of the reach range of the robot and it lets the IBVS task converge faster than the opposite repulsive pose.1

scholarly journals Uncertainty-Aware Knowledge Distillation for Collision Identification of Collaborative Robots

Sensors ◽  
2021 ◽  
Vol 21 (19) ◽  
pp. 6674
Author(s):  
Wookyong Kwon ◽  
Yongsik Jin ◽  
Sang Jun Lee
Keyword(s):  
Neural Network ◽  
Collision Detection ◽  
Deep Neural Network ◽  
Human Robot Interaction ◽  
Learning Method ◽  
Robot Interaction ◽  
Collaborative Robots ◽  
Knowledge Distillation ◽  
Articulated Robots ◽  
External Forces

Human-robot interaction has received a lot of attention as collaborative robots became widely utilized in many industrial fields. Among techniques for human-robot interaction, collision identification is an indispensable element in collaborative robots to prevent fatal accidents. This paper proposes a deep learning method for identifying external collisions in 6-DoF articulated robots. The proposed method expands the idea of CollisionNet, which was previously proposed for collision detection, to identify the locations of external forces. The key contribution of this paper is uncertainty-aware knowledge distillation for improving the accuracy of a deep neural network. Sample-level uncertainties are estimated from a teacher network, and larger penalties are imposed for uncertain samples during the training of a student network. Experiments demonstrate that the proposed method is effective for improving the performance of collision identification.

PEMFC Residual Life Prediction Using Sparse Autoencoder-Based Deep Neural Network

2019 ◽  
Vol 5 (4) ◽  
pp. 1279-1293 ◽  
Author(s):  
Jiawei Liu ◽  
Qi Li ◽  
Ying Han ◽  
Guorui Zhang ◽  
Xiang Meng ◽  
...  
Keyword(s):  
Neural Network ◽  
Life Prediction ◽  
Deep Neural Network ◽  
Residual Life ◽  
Sparse Autoencoder ◽  
Residual Life Prediction

Control of Autonomous Robots Using the Principles of Neuromodulation

2013 ◽  
Author(s):  
Akimul Prince ◽  
Biswanath Samanta
Keyword(s):  
Neural Network ◽  
Autonomous Robots ◽  
Neuronal Model ◽  
Autonomous Robot ◽  
Human Robot Interaction ◽  
Robot Interaction ◽  
Control Approach ◽  
The Neural Network ◽  
Simple Neural Network ◽  
Interesting Behavior

The paper presents a control approach based on vertebrate neuromodulation and its implementation on an autonomous robot platform. A simple neural network is used to model the neuromodulatory function for generating context based behavioral responses to sensory signals. The neural network incorporates three types of neurons — cholinergic and noradrenergic (ACh/NE) neurons for attention focusing and action selection, dopaminergic (DA) neurons for curiosity-seeking, and serotonergic (5-HT) neurons for risk aversion behavior. The implementation of the neuronal model on a relatively simple autonomous robot illustrates its interesting behavior adapting to changes in the environment. The integration of neuromodulation based robots in the study of human-robot interaction would be worth considering in future.

Gestures Identification for Myoelectric Based Control of the Robotic Arm

2020 ◽  
Vol 22 (12) ◽  
Author(s):  
Adhau P ◽  
◽  
Kadwane S. G ◽  
Shital Telrandhe ◽  
Rajguru V. S ◽  
...  
Keyword(s):  
Neural Network ◽  
Human Body ◽  
Human Robot Interaction ◽  
Robotic Arm ◽  
Human Hand ◽  
Robot Interaction ◽  
The Neural Network

Human robot interaction have been ever the topic of research to research scholars owing to its importance to help humanity. Robust human interacting robot where commands from Electromyogram (EMG) signals is recently being investigated. This article involves study of motions a system that allows signals recorded directly from a human body and thereafter can be used for control of a small robotic arm. The various gestures are recognized by placing the electrodes or sensors on the human hand. These gestures are then identified by using neural network. The neural network will thus train the signals. The offline control of the arm is done by controlling the motors of the robotic arm.

Open-circuit fault diagnosis of power rectifier using sparse autoencoder based deep neural network

2018 ◽  
Vol 311 ◽  
pp. 1-10 ◽  
Author(s):  
Lin Xu ◽  
Maoyong Cao ◽  
Baoye Song ◽  
Jiansheng Zhang ◽  
Yurong Liu ◽  
...  
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Deep Neural Network ◽  
Open Circuit ◽  
Sparse Autoencoder ◽  
Circuit Fault Diagnosis ◽  
Power Rectifier
Sign in / Sign up

Export Citation Format

Share Document