Tactile object recognition in early phases of grasping using underactuated robotic hands

Fonseca, Vinícius Prado da; Jiang, Xianta; Petriu, Emil M.; Oliveira, Thiago Eustaquio Alves de

doi:10.1007/s11370-022-00433-7

Cited by 12 publications

(7 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The module can also be miniaturized with custom PCBs designed to reduce its overall scale. Multiple instances of the module can be assembled on end-effectors’ fingers for in-hand object pose estimation [14] and identification [15] . The CAD files, microcontroller firmware, and Robot Operating System (ROS) [16] support the reproducibility of the module presented in this paper.…”

Section: Hardware In Contextmentioning

confidence: 99%

“…The articles also highlight the importance of considering the direction of exploration and the anisotropy of materials for texture recognition. Object recognition: In [15] , the authors propose a data-driven approach for object classification using underactuated hands and multimodal tactile sensors mounted on finger phalanges. The paper explores the utilization of tactile data in two scenarios: single-grasp, where the hand grasps an object quickly, and short-exploratory movements, where the hand disturbs the pose of an object to stimulate the sensors.…”

Section: Hardware Descriptionmentioning

confidence: 99%

“…The proposed tactile sensing apparatus is modularized to be embedded in the fingers or jaws of robotic end-effectors [14] , [15] , tactile probes [21] , or other robotic parts that may be susceptible to physical interactions in unstructured environments with external objects in planned or unplanned contact events. In terms of pressure, the proposed sensing module was characterized in [13] .…”

Section: Validation and Characterizationmentioning

confidence: 99%

“…In the context of the robotic hands assembled onto robotic hands, the data collected by the sensing module has been used in robotic tasks like stable grasping control [14] , in-hand object pose estimation, dynamic tactile texture recognition [19] , [20] , and object recognition [15] . A video of the sensors on the phalanx of underactuated fingers of a hybrid robotic hand can be found on: https://youtu.be/AcfX8dV5Sao .…”

Section: Validation and Characterizationmentioning

confidence: 99%

See 3 more Smart Citations

BioIn-Tacto: A compliant multi-modal tactile sensing module for robotic tasks

Alves de Oliveira,

Prado da Fonseca

2023

HardwareX

View full text Add to dashboard Cite

Section: Hardware In Contextmentioning

confidence: 99%

Section: Hardware Descriptionmentioning

confidence: 99%

Section: Validation and Characterizationmentioning

confidence: 99%

Section: Validation and Characterizationmentioning

confidence: 99%

See 2 more Smart Citations

BioIn-Tacto: A compliant multi-modal tactile sensing module for robotic tasks

Alves de Oliveira,

Prado da Fonseca

2023

HardwareX

View full text Add to dashboard Cite

“…An artificial neural network was used to identify five objects with the sensors’ information, achieving accuracy of 95%. In [ 33 ], a tactile sensor was located on the robotic fingers’ phalanges (two sensors per finger). The sensors measured the pressure, gravity, angular rate, and magnetic field.…”

Section: Introductionmentioning

confidence: 99%

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

Arco

Pontes

Segatto

et al. 2023

Sensors

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