• Journal of Sensor Science and Technology
• /
• v.15 no.4
• /
• pp.277-283
• /
• 2006

We presented that fabrication process and characteristics of 3 axes flexible tactile sensor available for normal and shear force fabricated using Si micromachining and packaging technologies. The fabrication processes for 3 axes flexible tactile sensor were classified in the fabrication of sensor chips and their packaging on the flexible PCB. The variation rate of resistance was about 2.1 %/N and 0.5 %/N in applying normal and shear force, respectively. The flexibility of fabricated 3 axes flexible tactile sensor array was good enough to place on the finger-tip.

• The Journal of Korea Robotics Society
• /
• v.17 no.3
• /
• pp.365-372
• /
• 2022

The accurate object recognition is important for the precise and accurate manipulation. To enhance the recognition performance, we can use various types of sensors. In general, acquired data from sensors have a high sampling rate. So, in the past, the RNN-based model is commonly used to handle and analyze the time-series sensor data. However, the RNN-based model has limitations of excessive parameters. CNN-based model also can be used to analyze time-series input data. However, CNN-based model also has limitations of the small receptive field in early layers. For this reason, when we use a CNN-based model, model architecture should be deeper and heavier to extract useful global features. Thus, traditional methods like RN N -based and CN N -based model needs huge amount of learning parameters. Recently studied result shows that Fast Fourier Convolution (FFC) can overcome the limitations of traditional methods. This operator can extract global features from the first hidden layer, so it can be effectively used for feature extracting of sensor data that have a high sampling rate. In this paper, we propose the algorithm to recognize objects using tactile sensor data and the FFC model. The data was acquired from 11 types of objects to verify our posed model. We collected pressure, current, position data when the gripper grasps the objects by random force. As a result, the accuracy is enhanced from 84.66% to 91.43% when we use the proposed FFC-based model instead of the traditional model.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.4
• /
• pp.1732-1739
• /
• 2018

With the explosion of digital devices, interaction technologies between human and devices are required more than ever. Especially, hand gesture recognition is advantageous in that it can be easily used. It is divided into the two groups: the contact sensor and the non-contact sensor. Compared with non-contact gesture recognition, the advantage of contact gesture recognition is that it is able to classify gestures that disappear from the sensor's sight. Also, since there is direct contacted with the user, relatively accurate information can be acquired. Electromyography (EMG) and force-sensitive resistors (FSRs) are the typical methods used for contact gesture recognition based on muscle activities. The sensors, however, are generally too sensitive to environmental disturbances such as electrical noises, electromagnetic signals and so on. In this paper, we propose a novel contact gesture recognition method based on Flexible Epidermal Tactile Sensor Array (FETSA) that is used to measure electrical signals according to movements of the wrist. To recognize gestures using FETSA, we extracted feature sets, and the gestures were subsequently classified using the support vector machine. The performance of the proposed gesture recognition method is very promising in comparison with two previous non-contact and contact gesture recognition studies.

• The Journal of Korea Robotics Society
• /
• v.4 no.1
• /
• pp.34-42
• /
• 2009

Rodents, specially rats, can recognize distance and shape of an object and also pattern of the textures by using their whiskers. Mechanoreceptors surrounding the root of whisker in their follicle measure deflection of the whisker. Rats can move their whisker back and forth freely. This ability, called active whisking or active sensing, is one of characteristics of rat behaviours. Many researches based on the mechanism have been progressed. In this paper, we test a simple and accurate method based on deflection of the whisker: we designed biomimetic whiskers modeling after a structure of follicle using the microphone. The microphone sensor measures a mechanical vibration. Attaching an artificial whisker beam to the microphone membrane, we can detect a vibration of whisker and this can show the deflection amount of whisker indirectly.

• Proceedings of the KIEE Conference
• /
• 2007.04a
• /
• pp.352-354
• /
• 2007

This paper presents a method to estimate the position of a mobile robot by using a gyro sensor and accelerometer sensors on it. Together with contact sensors we propose a mapmaking algorithm for an indoor environment where the robot moves. The direction of robot can be estimated through a gyro sensor and the distance is founded out by accelerometers. Then one can presume the position of robot. Using the direction and distance values vector-based mapmaking job can be performed. Tactile sensors help the robot recognize the boundary limit value of indoor environment and decide outer wall line of the map.

• Electronics and Telecommunications Trends
• /
• v.34 no.2
• /
• pp.10-18
• /
• 2019

Since the introduction of CRT(Cathode Ray Tube) in the 1950s, display technologies have been developed continuously. Flat panel displays such as PDP(Plasma Display Panel) and LCD(Liquid Crystal Display) were commercialized in the late 1990s, and OLED(Organic Light Emitting Diodes) and Micro-LED(Micro-Light Emitting Diodes) are now being developed and are becoming widespread. In the future, we expect to develop ultra-realistic, flexible, embedded sensor displays. Ultra-realistic display can be applied to AR/VR(Augmented Reality/Virtual Reality) devices and spatial light modulators for holography. The sensor-embedded display can be applied to robots; electronic skin; and security devices, including iris recognition sensors, fingerprint recognition sensors, and tactile sensors. AR/VR technology must be developed to meet technical requirements such as viewing angle, resolution, and refresh rate. Holography requires optical modulation technology that can significantly improve resolution, viewing angle, and modulation method to enable wide-view and high-quality hologram stereoscopic images. For electronic skin, stable mass production technology, large-area arrays, and system integration technologies should be developed.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2013.05a
• /
• pp.873-874
• /
• 2013

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.29 no.4 s.235
• /
• pp.540-554
• /
• 2005

This paper presents a motion planning and control method for the dexterous manipulation with a robotic hand. For a given trajectory of an object, a simulation system calculates the necessary joint displacements and contact forces at the fingertip surfaces. These joint displacements and contact forces are the reference inputs to the control loops of the robotic fingers. A task is decomposed into a set of primitive motions, and each primitive motion is executed using the planned output of the simulation system as the reference. Force sensors and dynamic tactile sensors are used to adapt to errors and uncertainties encountered during manipulation. Several experimental results are presented.

• The Journal of Korea Robotics Society
• /
• v.19 no.3
• /
• pp.266-273
• /
• 2024

With the rising interaction between robots and humans, detecting collisions has become increasingly vital for ensuring safety. In this paper, we propose a novel approach for detecting collisions without using force torque sensors or tactile sensors, utilizing a Transformer-based neural network architecture. The proposed collision detection approach comprises a torque estimator network that predicts the joint torque in a free-motion state using Synchronous time-step encoding, and a collision discriminator network that predicts collisions by leveraging the difference between estimated and actual torques. The collision discriminator finally creates a binary tensor that predicts collisions frame by frame. In simulations, the proposed network exhibited enhanced collision detection performance relative to the other kinds of networks both in terms of prediction speed and accuracy. This underscores the benefits of using Transformer networks for collision detection tasks, where rapid decision-making is essential.

• Journal of the Society of Naval Architects of Korea
• /
• v.52 no.6
• /
• pp.478-484
• /
• 2015

Knowledge about ice load distribution along the ship hull due to ship-ice interaction can provide important background information for the development of design codes for ice-going vessels. The objective of this study is to understand ship and ice interaction phenomena and determine the magnitude of ice load acting along a ship hull. The model tests were performed in the ice model basin in Korea Research Institute of Ships and Ocean engineering (KRISO) with the model of icebreaking ship Araon. Self-propulsion tests in level ice were performed with three difference model ship speeds. In the model tests, three tactile sensors were installed to measure the spatial distribution of ice load acting at different locations on a model ship, such as the bow and shoulder areas. Variation in the distribution of ice load acting on a model hull with ship speed is discussed.