scholarly journals Open Database for Accurate Upper-Limb Intent Detection Using Electromyography and Reliable Extreme Learning Machines

Sensors ◽  
2019 ◽  
Vol 19 (8) ◽  
pp. 1864 ◽  
Author(s):  
Vinicius Cene ◽  
Mauricio Tosin ◽  
Juliano Machado ◽  
Alexandre Balbinot
Keyword(s):  
Signal Processing ◽  
Signal Classification ◽  
Disruptive Technology ◽  
Extreme Learning Machines ◽  
Real Time Control ◽  
Semg Signal ◽  
Human Interface ◽  
Learning Machines ◽  
Time Control ◽  
Control Interface

Surface Electromyography (sEMG) signal processing has a disruptive technology potential to enable a natural human interface with artificial limbs and assistive devices. However, this biosignal real-time control interface still presents several restrictions such as control limitations due to a lack of reliable signal prediction and standards for signal processing among research groups. Our paper aims to present and validate our sEMG database through the signal classification performed by the reliable forms of our Extreme Learning Machines (ELM) classifiers, used to maintain a more consistent signal classification. To perform the signal processing, we explore the use of a stochastic filter based on the Antonyan Vardan Transform (AVT) in combination with two variations of our Reliable classifiers (denoted R-ELM and R-Regularized ELM (RELM), respectively), to derive a reliability metric from the system, which autonomously selects the most reliable samples for the signal classification. To validate and compare our database and classifiers with related papers, we performed the classification of the whole of Databases 1, 2, and 6 (DB1, DB2, and DB6) of the NINAProdatabase. Our database presented consistent results, while the reliable forms of ELM classifiers matched or outperformed related papers, reaching average accuracies higher than 99 % for the IEEdatabase, while average accuracies of 75 . 1 % , 79 . 77 % , and 69 . 83 % were achieved for NINAPro DB1, DB2, and DB6, respectively.

scholarly journals Moldless PEGDA-Based Optoelectrofluidic Platform for Microparticle Selection

2011 ◽  
Vol 2011 ◽  
pp. 1-8 ◽  
Author(s):  
Shih-Mo Yang ◽  
Tung-Ming Yu ◽  
Ming-Huei Liu ◽  
Long Hsu ◽  
Cheng-Hsien Liu
Keyword(s):  
Material Design ◽  
Real Time Control ◽  
Glycol Diacrylate ◽  
Single Spin ◽  
Time Control ◽  
Control Interface ◽  
Microfluidic Flow ◽  
Ito Glass ◽  
New Material ◽  
Poly Ethylene

This paper reports on an optoelectrofluidic platform which consists of the organic photoconductive material, titanium oxide phthalocyanine (TiOPc), and the photocrosslinkable polymer, poly (ethylene glycol) diacrylate (PEGDA). TiOPc simplifies the fabrication process of the optoelectronic chip due to requiring only a single spin-coating step. PEGDA is applied to embed the moldless PEGDA-based microchannel between the top ITO glass and the bottom TiOPc substrate. A real-time control interface via a touch panel screen is utilized to select the target 15 μm polystyrene particles. When the microparticles flow to an illuminating light bar, which is oblique to the microfluidic flow path, the lateral driving force diverts the microparticles. Two light patterns, the switching oblique light bar and the optoelectronic ladder phenomenon, are designed to demonstrate the features. This work integrating the new material design, TiOPc and PEGDA, and the ability of mobile microparticle manipulation demonstrates the potential of optoelectronic approach.

scholarly journals Robotic Walker for Slope Mobility Assistance with Active-Passive Hybrid Actuator

2021 ◽  
Vol 34 (1) ◽  
Author(s):  
Junqiang Li ◽  
Lei Zhao ◽  
Tiejun Li
Keyword(s):  
Mechanical Properties ◽  
Signal Processing ◽  
Control Algorithm ◽  
Control Method ◽  
Dc Motor ◽  
Real Time Control ◽  
Time Control ◽  
Robotic Walker ◽  
And Control ◽  
Hybrid Actuator

AbstractWalking assistance can be realized by active and passive robotic walkers when their users walk on even roads. However, fast signal processing and real-time control are necessary for active robotic walkers when the users walk on slopes, while assistive forces cannot be provided by passive robotic walkers when the users walk uphill. A robotic walker with an active-passive hybrid actuator (APHA) was developed in this study. The APHA, which consists of a rotary magnetorheological (MR) brake and a DC motor, can provide mobility assistance to users walking both uphill and downhill via the cooperative operation of the MR brake and DC motor. The rotary MR brake was designed with a T-shaped configuration, and the system was optimized to minimize the brake volume. Prototypes of the APHA and robotic walker were constructed. A control algorithm for the robotic walker was developed based on the characteristics of the APHA and the structure of the robotic walker. The mechanical properties of the APHA were characterized, and experiments were conducted to evaluate the mobility assistance supplied by the robotic walker on different roads. The results show that the APHA can meet the requirements of the robotic walker, and suitable assistive forces can be provided by the robotic walker, which has a simple mechanical structure and control method.

Sign in / Sign up

Export Citation Format

Share Document