scholarly journals Motion and Force Estimation System of Human Fingers

2011 ◽  
Vol 17 (10) ◽  
pp. 1014-1020 ◽  
Author(s):  
Dong-Chul Lee ◽  
Young-Jin Choi
Author(s):  
Soojin Cho ◽  
Jerome Peter Lynch ◽  
Chung-Bang Yun

Cable tension force is one of the most important structural parameters to monitor in cable-stayed bridges. For example, cable tension needs to be monitored during construction and maintenance to ensure the bridge is not overloaded. To economically monitor tension forces, this study proposes the use of an automated wireless tension force estimation system (WFTES) developed solely for cable force estimation. The design of the WFTES system can be divided into two parts: low-cost hardware and automated software. The low-cost hardware consists of an integrated platform containing a wireless sensing unit constructed from commercial off-the-shelf components, a low-cost commercial MEMS accelerometer, and a signal conditioning board for signal amplification and filtering. With respect to the automated software, a vibration-based algorithm using estimated modal parameters and information on the cable sag and bending stiffness is embedded into the wireless sensing unit. Since modal parameters are inputs to the algorithm, additional algorithms are necessary to extract modal features from measured cable accelerations. To validate the proposed WFTES, a scaled-down cable model was constructed in the laboratory using steel rope wire. The wire was exposed to broad-band excitations while the WFTES recorded the cable response and embedded algorithms interrogated the measured acceleration to estimate tension force. The results reveal the embedded algorithms properly identify the lower natural frequencies of the cable and make accurate estimates of cable tension. This paper concludes with a summary of the salient research findings and suggestions for future work.


1996 ◽  
Vol 8 (3) ◽  
pp. 226-234
Author(s):  
Kiyoshi Ohishi ◽  
◽  
Masaru Miyazaki ◽  
Masahiro Fujita ◽  

Generally, hybrid control is realized by sensor signal feedback of position and force. However, some robot manipulators do not have a force sensor due to the environment. Moreover, a precise force sensor is very expensive. In order to overcome these problems, we propose the estimation system of reaction force without using a force sensor. This system consists of the torque observer and the inverse dynamics calculation. Using both this force estimation system and <I>H</I>∞ acceleration controller which is based on <I>H</I>∞ control theory, it takes into account the frequency characteristics of both sensor noise effect and disturbance rejection. The experimental results in this paper illustrate the fine hybrid control of the three tested degrees-of-freedom DD robot manipulator without force sensor.


2021 ◽  
Vol 33 (6) ◽  
pp. 1349-1358
Author(s):  
Yoshiyuki Higashi ◽  
◽  
Kenta Yamazaki ◽  
Arata Masuda ◽  
Nanako Miura ◽  
...  

This paper presents an attractive force estimation system and an automatic activation system for an electropermanent magnet (EPM) for an inspection UAV. Adsorption to infrastructures for inspection at a distance is extremely difficult to perform safely because the operator cannot detect the state of adsorption of the drone equipped with a magnetic adsorption device. Therefore, in this paper, we clarify the relationship between the magnetic flux density and attractive force of the EPM through experiments, and develop an estimation algorithm for the attractive force based on the results. An automatic activation system, using the induced voltage in the coil when the EPM approaches the magnetic substance, is developed and mounted on a quadrotor for a flight experiment along with the estimation system for the attractive force. The developed system is verified using flight and adsorption experiments on the quadrotor.


Author(s):  
Shiva Nosouhian ◽  
Fereshteh Nosouhian ◽  
Abbas Kazemi Khoshouei

Deep neural networks (DNNs) have made a huge impact in the field of machine learning by providing unbeatable humanlike performance to solve real-world problems such as image processing and natural language processing (NLP). Convolutional neural network (CNN) and recurrent neural network (RNN) are two typical architectures that are widely used to solve such problems. Time sequence-dependent problems are generally very challenging, and RNN architectures have made an enormous improvement in a wide range of machine learning problems with sequential input involved. In this paper, different types of RNN architectures are compared. Special focus is put on two well-known gated-RNN&rsquo;s Long Term Short Memory (LSTM) and Gated Recurrent Unit (GRU). We evaluated these models on the task of force estimation system in pouring. In this study, four different models including multi-layers LSTM, multi-layers GRU, single-layer LSTM and single-layer GRU) were created and trained. The result suggests that multi-layer GRU outperformed other three models.


2008 ◽  
Vol 2008.46 (0) ◽  
pp. 277-278
Author(s):  
Takashi OCHI ◽  
Liu TAO ◽  
Yoshio INOUE ◽  
Kyoko SHIBATA

Sign in / Sign up

Export Citation Format

Share Document