Optical Fiber Angle Sensors for the PrHand Prosthesis: Development and Application in Grasp Types Recognition with Machine Learning

Arco, Laura De; Pontes, María José; Segatto, Marcelo E. V.; Monteiro, Maxwell E.; Cifuentes, Carlos A.; Díaz, Camilo A. R.

doi:10.1109/laedc54796.2022.9908232

Cited by 6 publications

(7 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sensor Testing The experiments show evident variations between the different objects, hence, the use of these data for machine learning algorithms is good. The angle sensors were already tested in a KNN classifier, showing a good algorithm response with an accuracy of around 93 % [19]. This is confirmed by the signals response shown in Figure 5, where the difference between the objects is observed.…”

Section: 3supporting

confidence: 55%

“…In [17], the prosthesis functionally was assessed for which grip types the prosthesis better performs, comparing to the grasp types the human hand performs during activities of daily life, namely: hook, spherical grip, tripod pinch, and spherical grip. In [19] the angle sensors information was used for grasp types recognition with machine learning (ML) algorithms, taking into account the grasp types aforementioned. Thus, in this study the same four objects that were used in [19] by grip type are used.…”

Section: Sensor Testingmentioning

confidence: 99%

“…In [19] the angle sensors information was used for grasp types recognition with machine learning (ML) algorithms, taking into account the grasp types aforementioned. Thus, in this study the same four objects that were used in [19] by grip type are used. However, at this time both the angle and the pressure sensors are analyzed under grasping tasks.…”

Section: Sensor Testingmentioning

confidence: 99%

See 2 more Smart Citations

Pressure and Angle Sensors with Optical Fiber for Instrumentation of the PrHand Hand Prosthesis

Arco

Pontes

Segatto

et al. 2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

The principal cause of upper limb amputations is due to traumatism. The prosthesis is an assistive device to help in the activities of daily for the amputee person. However, one of the latest reports shows that in developing countries there are around 30 million people without assistive devices. This work presents the development of two kinds of sensors for the PrHand, an upper limb prosthesis based on compliant mechanism and soft-robotics. The sensors are made with polymeric optical fiber (POF), due to their flexibility and low cost, and the working principle is based on intensity variation. The angle sensors are used for monitoring the interphalangeal joint of the fingers, and for the assessment were made cycles of closing and opening each finger. On the other hand, the force sensors are located at the tip of three fingers to track the force made over the objects. Before encoring the sensors were evaluated making five cycles of compressing and decompressing each sensor. The results show a linear behavior between the angle and the voltage variation, one most remarkable angle sensor result was with a sensibility of 0.0357 V/° and an R2 of 99 % closing and 0.0483 V/° opening. In the case of the force sensor, a polynomial relation was found between the voltage changes and the pressure over the sensor; in some cases, the relation between voltage changes and pressure could be linear but that depends on the construction of the sensor. Regarding the obtained R2 of 99 %, its sensibility was 0.0361 V/N compression and 0.0368 V/N decompression. In conclusion, was successfully developed two kinds of sensors for the instrumentation of PrHand prosthesis. It is expected to use angle and sensor variables as input in algorithms of Machine Learning to improve the detection of objects. One aspect to improve is to control in a better way the sensor construction parameters due to the considerable influence over the sensor behavior.

show abstract

Section: 3supporting

confidence: 55%

Section: Sensor Testingmentioning

confidence: 99%

Section: Sensor Testingmentioning

confidence: 99%

See 1 more Smart Citation

Pressure and Angle Sensors with Optical Fiber for Instrumentation of the PrHand Hand Prosthesis

Arco

Pontes

Segatto

et al. 2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

show abstract

“…In a preliminary study, the angle sensors with side polish were used to instrument the PrHand prosthesis, and their information was used to recognize four of the eight grasp types (H, SG, TP, and CG). The k-NN algorithm was used in the study, and the k-fold accuracy result was 92.81 ± 0.47% [ 48 ]. Comparing the results of this study with the last one, this study had better results since the accuracy was 98.5 ± 0.01%.…”

Section: Resultsmentioning

confidence: 99%

“…The characterization of the first sensor (side polish) was described in previous works [ 47 , 48 , 49 ]. This sensor was already used to instrument the PrHand prosthesis, and it is used in this study for comparison purposes.…”

Section: Methodsmentioning

confidence: 99%

Soft-Sensor System for Grasp Type Recognition in Underactuated Hand Prostheses

Arco

Pontes

Segatto

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

This paper presents the development of an intelligent soft-sensor system to add haptic perception to the underactuated hand prosthesis PrHand. Two sensors based on optical fiber were constructed, one for finger joint angles and the other for fingertips’ contact force. Three sensor fabrications were tested for the angle sensor by axially rotating the sensors in four positions. The configuration with the most similar response in the four rotations was chosen. The chosen sensors presented a polynomial response with R2 higher than 92%. The tactile force sensors tracked the force made over the objects. Almost all sensors presented a polynomial response with R2 higher than 94%. The system monitored the prosthesis activity by recognizing grasp types. Six machine learning algorithms were tested: linear regression, k-nearest neighbor, support vector machine, decision tree, k-means clustering, and hierarchical clustering. To validate the algorithms, a k-fold test was used with a k = 10, and the accuracy result for k-nearest neighbor was 98.5%, while that for decision tree was 93.3%, enabling the classification of the eight grip types.

show abstract

Single‐mode optical fibre digital decoder based on polarization using a K‐nearest neighbour algorithm

Ibáñez‐Camarillo,

Martínez‐Piñón,

Márquez‐Olivera

et al. 2023

Electronics Letters

View full text Add to dashboard Cite

It is shown experimentally a new digital optical decoding scheme based on the transmission or polarized light at p polarization planes using a K‐Nearest Neighbour (KNN) algorithm through a single‐mode optical fibre at 633 nm. The optical power signal is sent at p polarization planes which constitute p classes required for signal bit recognition. Results show that it is possible to recognize 32 polarizations planes, 5 bits, using 4 features corresponding to the measurement of optical power at 4 different angles at the photodetector side with an average assertiveness of 99.02%.

show abstract

Optical Fiber Angle Sensors for the PrHand Prosthesis: Development and Application in Grasp Types Recognition with Machine Learning

Cited by 6 publications

References 17 publications

Pressure and Angle Sensors with Optical Fiber for Instrumentation of the PrHand Hand Prosthesis

Pressure and Angle Sensors with Optical Fiber for Instrumentation of the PrHand Hand Prosthesis

Soft-Sensor System for Grasp Type Recognition in Underactuated Hand Prostheses

Single‐mode optical fibre digital decoder based on polarization using a K‐nearest neighbour algorithm

Contact Info

Product

Resources

About