Surface Covering Structure and Active Sensing with MEMS-CMOS Integrated 3-Axis Tactile Sensors for Object Slip Detection and Texture Recognition

2021 ◽  
Author(s):  
Sumeyya Javaid ◽  
Hideki Hirano ◽  
Shuji Tanaka ◽  
Masanori Muroyama
Keyword(s):  
Active Sensing ◽  
Tactile Sensors ◽  
Texture Recognition ◽  
Surface Covering ◽  
Slip Detection

scholarly journals Biomimetic Tactile Sensors with Bilayer Fingerprint Ridges Demonstrating Texture Recognition

Micromachines ◽  
2019 ◽  
Vol 10 (10) ◽  
pp. 642 ◽  
Author(s):  
Eunsuk Choi ◽  
Onejae Sul ◽  
Jusin Lee ◽  
Hojun Seo ◽  
Sunjin Kim ◽  
...  
Keyword(s):  
Elastic Moduli ◽  
Tactile Sensor ◽  
Skin Layer ◽  
Surface Structures ◽  
Tactile Sensitivity ◽  
Detection Capability ◽  
Tactile Sensors ◽  
Texture Recognition ◽  
Shear Transfer ◽  
Geometrical Factors

In this article, we report on a biomimetic tactile sensor that has a surface kinetic interface (SKIN) that imitates human epidermal fingerprint ridges and the epidermis. The SKIN is composed of a bilayer polymer structure with different elastic moduli. We improved the tactile sensitivity of the SKIN by using a hard epidermal fingerprint ridge and a soft epidermal board. We also evaluated the effectiveness of the SKIN layer in shear transfer characteristics while varying the elasticity and geometrical factors of the epidermal fingerprint ridges and the epidermal board. The biomimetic tactile sensor with the SKIN layer showed a detection capability for surface structures under 100 μm with only 20-μm height differences. Our sensor could distinguish various textures that can be easily accessed in everyday life, demonstrating that the sensor may be used for texture recognition in future artificial and robotic fingers.

Extracting Multimodal Dynamics of Objects Using RNNPB

2005 ◽  
Vol 17 (6) ◽  
pp. 681-688 ◽  
Author(s):  
Tetsuya Ogata ◽  
◽  
Hayato Ohba ◽  
Jun Tani ◽  
Kazunori Komatani ◽  
...  
Keyword(s):  
Neural Network ◽  
Recurrent Neural Network ◽  
Active Sensing ◽  
Dynamic Feature ◽  
Extended Version ◽  
Dynamic Features ◽  
Tactile Sensors ◽  
Self Organized ◽  
Using Data

Dynamic features play an important role in recognizing objects that have similar static features in color or shape. This paper focuses on active sensing that exploits the dynamic feature of an object. An extended version of the robot, Robovie-IIs, uses its arms to move an object and determine its dynamic features. At issue is how to extract symbols from different temporal states of the object. We use a <I>recurrent neural network with parametric bias</I> (RNNPB) that generates self-organized nodes in parametric bias space. We trained an RNNPB with 42 neurons using data on sounds, trajectories, and tactile sensors generated while the robot was moving or hitting an object with its arm. Clusters of 20 types of objects were self-organized. Experiments with unknown (untrained) objects showed that our proposal configured them appropriately in PB space, demonstrating its <I>generalization</I>.

Bond slip detection of concrete-encased composite structure using shear wave based active sensing approach

2015 ◽  
Vol 24 (12) ◽  
pp. 125026 ◽  
Author(s):  
Lei Zeng ◽  
Seyed Mohammad Parvasi ◽  
Qingzhao Kong ◽  
Linsheng Huo ◽  
Ing Lim ◽  
...  
Keyword(s):  
Shear Wave ◽  
Composite Structure ◽  
Active Sensing ◽  
Bond Slip ◽  
Slip Detection

A New Highly Sensitive Dielectric Slip Detection Sensor for a Robotic Operation

2017 ◽  
Author(s):  
Sung Joon Kim ◽  
Ja Choon Koo
Keyword(s):  
Tactile Sensor ◽  
Relative Displacement ◽  
Large Area ◽  
Tactile Sensors ◽  
Post Process ◽  
Highly Sensitive ◽  
Grasping And Manipulation ◽  
Working Principle ◽  
Slip Detection

For dexterous grasping and manipulation, tactile sensors recognizing contact object are essential. Electronic skin (E-skin) with tactile sensors plays a role as both receiving information for grasping and protecting robot frame. This paper presents a polymer tactile sensor covering large area to fulfill role of E-skin. The sensor has a thin air gap between polymer layers and it is deformed reacting slip input. When slip is occurred, there is relative displacement between surrounding layer and it incurs change of electrode separation. NBR is used to sensor substrate because of its tough and flexible characteristic. Ultrathin aluminum tape is employed for electrodes. There is a changeability of size of the sensor because of its simple but effective working principle and structure. Slip detecting algorithm doesn’t have a post process such as FFT or DWT, so there isn’t delay for processing time. It realizes real-time slip detection reducing reaction time of robot hand.

scholarly journals Deep Vibro-Tactile Perception for Simultaneous Texture Identification, Slip Detection, and Speed Estimation

Sensors ◽  
2020 ◽  
Vol 20 (15) ◽  
pp. 4121
Author(s):  
Yerkebulan Massalim ◽  
Zhanat Kappassov ◽  
Huseyin Atakan Varol
Keyword(s):  
Neural Networks ◽  
Short Term Memory ◽  
Dexterous Manipulation ◽  
Texture Recognition ◽  
Contact Points ◽  
Tactile Interaction ◽  
Feed Forward Neural Networks ◽  
Long Short Term Memory ◽  
Generalization Accuracy ◽  
Slip Detection

Autonomous dexterous manipulation relies on the ability to recognize an object and detect its slippage. Dynamic tactile signals are important for object recognition and slip detection. An object can be identified based on the acquired signals generated at contact points during tactile interaction. The use of vibrotactile sensors can increase the accuracy of texture recognition and preempt the slippage of a grasped object. In this work, we present a Deep Learning (DL) based method for the simultaneous texture recognition and slip detection. The method detects non-slip and slip events, the velocity, and discriminate textures—all within 17 ms. We evaluate the method for three objects grasped using an industrial gripper with accelerometers installed on its fingertips. A comparative analysis of convolutional neural networks (CNNs), feed-forward neural networks, and long short-term memory networks confirmed that deep CNNs have a higher generalization accuracy. We also evaluated the performance of the highest accuracy method for different signal bandwidths, which showed that a bandwidth of 125 Hz is enough to classify textures with 80% accuracy.

Development of Direct-printed Tactile Sensors for Gripper Control through Contact and Slip Detection

2018 ◽  
Vol 16 (2) ◽  
pp. 929-936 ◽  
Author(s):  
Ju-Kyoung Lee ◽  
Hyun-Hee Kim ◽  
Jae-Won Choi ◽  
Kyung-Chang Lee ◽  
Suk Lee
Keyword(s):  
Tactile Sensors ◽  
Slip Detection

scholarly journals Tactile Sensors for Friction Estimation and Incipient Slip Detection—Toward Dexterous Robotic Manipulation: A Review

2018 ◽  
Vol 18 (22) ◽  
pp. 9049-9064 ◽  
Author(s):  
Wei Chen ◽  
Heba Khamis ◽  
Ingvars Birznieks ◽  
Nathan F. Lepora ◽  
Stephen J. Redmond
Keyword(s):  
Robotic Manipulation ◽  
Tactile Sensors ◽  
Friction Estimation ◽  
Slip Detection
Sign in / Sign up

Export Citation Format

Share Document