• The Journal of Korea Robotics Society
• /
• v.19 no.1
• /
• pp.16-22
• /
• 2024

Robotic grasping in unstructured environments poses a significant challenge, demanding precise estimation of gripping positions for diverse and unknown objects. Generative Grasping Convolution Neural Network (GG-CNN) can estimate the position and direction that can be gripped by a robot gripper for an unknown object based on a three-dimensional depth map. Since GG-CNN uses only a depth map as an input, the precision of the depth map is the most critical factor affecting the result. To address the challenge of depth map precision, we integrate the Segment Anything Model renowned for its robust zero-shot performance across various segmentation tasks. We adjust the components corresponding to the segmented areas in the depth map aligned through external calibration. The proposed method was validated on the Cornell dataset and SurgicalKit dataset. Quantitative analysis compared to existing methods showed a 49.8% improvement with the dataset including surgical instruments. The results highlight the practical importance of our approach, especially in scenarios involving thin and metallic objects.

• The Journal of Korea Robotics Society
• /
• v.19 no.3
• /
• pp.274-280
• /
• 2024

With the recent emergence of collaborative robots, demand for grippers to be used in service sites is increasing. Soft grippers have advantages in terms of compliance, cost, weight, and stability, but it is difficult to handle objects of various shapes, sizes, and weights with a single gripper. In this study, based on a fluidic elastomer actuator (FEA), we propose a pneumatic soft gripper that can grip a variety of objects, consisting of fingers capable of both grasping and vacuum suction and a base with variable stroke between the fingers using a pneumatic cylinder. A check valve is installed inside the finger, so that when positive pressure is supplied, the finger expands to perform adaptive grasping, and when negative pressure is supplied, vacuum suction can be performed. Grasping experiments were conducted on various objects to evaluate the performance of the proposed gripper, and 94% of 54 gripping objects were successfully grasped.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.3
• /
• pp.1229-1237
• /
• 2012

This study aimed to investigate the effect of hand grip force on the activity of shoulder muscles in 2 arm postures. The muscle activity of the upper trapezius, lower trapezius, infraspinatus, anterior deltoid, and posterior deltoid muscles in 22 healthy subjects (11 men and 11 women) were measured using surface electrodes during 4 hand gripping tasks (0%, 30%, 50%, and 70% of maximum voluntary contraction) in 2 shoulder positions (neutral position and $90^{\circ}$ elevation position). Among changing grip force significantly differenced infraspinatus, anterior deltoid, and posterior deltoid muscles' activity in the shoulder neutral position(p<0.05). In the shoulder $90^{\circ}$ elevation position, anterior deltoid, posterior deltoid muscles' activity was significantly differenced(p<0.05). Hand gripping was found to alter muscle activation. The hand grip task activated the infraspinatus muscle in the neutral position and inhibited the deltoid muscle in the $90^{\circ}$ elevation position. This finding may prove useful for the development preventative measures and rehabilitation strategies for shoulder injuries.

• Transactions on Electrical and Electronic Materials
• /
• v.13 no.5
• /
• pp.215-220
• /
• 2012

Miniature ultrasonic and tactile sensors on Si substrate have been proposed, fabricated and characterized to detect objects for a dexterous robot. The ultrasonic sensor consists of piezoelectric PZT thin film on a Pt/Ti/$SiO_2$ and/or Si diaphragm fabricated using a micromachining technique; the ultrasonic sensor detects the piezoelectric voltage as an ultrasonic wave. The sensitivity has been enhanced by improving the device structure, and the resonant frequency in the array sensor has been equalized. Position detection has been carried out by using a sensor array with high sensitivity and uniform resonant frequency. The tactile sensor consists of four or three warped cantilevers which have NiCr or $Si:B^+$ piezoresistive layer for stress detection. Normal and shear stresses can be estimated by calculation using resistance changes of the piezoresitive layers on the cantilevers. Gripping state has been identified by using the tactile sensor which is installed on finger of a robot hand, and friction of objects has been measured by slipping the sensor.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.423-427
• /
• 2003

In this paper, it is to development of brake system with ABS function for aircraft. The test of brake system is required before applying on aircraft. The real-time dynamic simulator with 5-D.O.F. aircraft dynamic model is developed for braking performance test of ABS (Anti-skid Brake System) control h/w with anti-skid brake functions. The dynamic simulator is real-time interface system that is composed of dynamic simulation parts, master control parts, digital and analog in/out interface parts, and user interface parts. The 5-D.O.F. aircraft dynamic model is composed of a big contour and a little contour by simulation s/w. The big contour represents the interactions of forces in airframe, nose and main landing gear, and engines on the center of gravity. The little contour represents interactions of wheel, braking units, hydraulic units and a control unit. ABS control h/w unit with ABS control algorithm is also developed and is tested with simulator under the some conditions of gripping coefficient. We have known that ABS control h/w unit on wet or snowy runway as well as dry runway very well protects wheel skid.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10b
• /
• pp.202-207
• /
• 1993

A theoretical and experimental study is presented for the force holding control of a miniature robotic ringer which is driven by a pair of piezoelectric unimorph cells. In the theoretical analysis, one finger is modeled as a flexible cantilever with a tactile force sensor at the tip and the mate of the finger is a solid beam supposed with sufficient stiffness. Further, the force sensor is modeled by a one-degree-of-freedom, mass-spring system and the output of sensor is then described by the sensor stiffness multiplied by the relative displacement. The problem investigated in this paper is that two typical holding tasks of the human finger are picked up and applied to the robotic finger. One is the work holding a stationary object with a prescribed, time-varying force and the other one is to keep the contacted force constant even if the object is in motion. The simple PID feedback control scheme is used to control the minute gripping force of order 0.01 Newton. It is shown both experimentally and theoretically that the artificial finger with the piezoelectric actuator works well in the minute force holding of the tiny object.

• Journal of Biosystems Engineering
• /
• v.37 no.1
• /
• pp.65-74
• /
• 2012

Purpose: An autonomous robot was developed for harvesting strawberries cultivated in bench-type systems. Methods: The harvest robot consisted of four main components: an autonomous vehicle, a manipulator with four degrees of freedom (DOF), an end effector with two DOFs, and a color computer vision system. Strawberry detection was performed based on 3D image and distance information obtained from a stereo CCD color camera and a laser device, respectively. Results: In this work, a Cartesian type manipulator system was designed, including an intermediate revolute axis and a double driven arm-based joint axis, so that it could generate collision-free motions during harvesting. A DC servomotor-driven end-effector, consisting of a gripper and a cutter, was designed for gripping and cutting the strawberry stem without damaging the strawberry itself. Real-time position tracking algorithms were developed to detect, recognize, trace, and approach strawberries under natural light conditions. Conclusion: The developed robot system could harvest a strawberry within 7 seconds without damage.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.15 no.6
• /
• pp.108-113
• /
• 2007

This study aims to find equivalent comfort contours, reciprocal of frequency weighting curves, for vertical steering wheel vibration. Psychophysical responses were measured from twelve male subjects by using magnitude estimation of relative discomfort due to vertical steering wheel vibrations of magnitude of 0.1 to 1.58 $m/s^2$ in the frequency range of 4 to 250 Hz. Relative discomfort were estimated with a reference vibration of 0.4 $m/s^2$ at 31.5 Hz. Equivalent comfort contours were produced from the median of sensation magnitudes judged by twelve subjects, which showed variation in the shapes with increase of vibration magnitude. A shape of the contour came close to the perception threshold curve with decrease of vibration magnitude. When the vibration magnitude increases, the shape changed close to those in the references of Hong and et al (2003). It is also recommended frequency weighting curves for vertical steering wheel vibration must be expressed as a function of vibration magnitude as well as frequency.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.16 no.12
• /
• pp.2315-2327
• /
• 1992

A robot assembly method which uses configuration and force/torque information of tactile sensor system and performs chamferless peg-in-hole tasks is suggested and experimentally studied. When the robot gripes the peg with random orientation, the realignment of the peg to the hole center line is successfully performed with the gripping configuration information of the tactile sensor and the inverse kinematics of the robot. The force/torque information of the tactile sensor makes it possible to control the contacting force between mating parts during hole search stage. The suggested algorithm employs a hybrid position/force control and the experiments show that the algorithm accomplishes well peg-in-hole tasks with permissible small contacting force. The chamferless peg-in-hole tasks with smaller clearance than the robot repeatibility can be excuted without any loss or deformation of mating parts. This study the possibility of precise and chamferless parts mating by robot and tactile sensor system.

• Journal of the Korean Society for Precision Engineering
• /
• v.20 no.7
• /
• pp.105-113
• /
• 2003

This paper describes the development of a 6-axis robot's finger force/moment sensor, which measures forces Fx, Fy, Fz, and moments Mx, My, Mz simultaneously, for stably grasping an unknown object. In order to safely grasp an unknown object using the robot's gripper, it should measure the force in the gripping direction and the force in the gravity direction, and perform the force control using the measured forces. Thus, the robot's gripper should be composed of 6-axis robot's finger force/moment sensor that can measure forces Fx, Fy, Fz, and moments Mx, My, Mz simultaneously. In this paper, the 6-axis robot's finger force/moment sensor for measuring forces Fx, Fy, Fz, and moments Mx, My, Mz simultaneously was newly modeled using several parallel-plate beams, designed, and fabricated. The characteristic test of made sensor was performed. and the result shows that interference errors of the developed sensor are less than 3%. Also, Robot's gripper with the 6-axis robot's finger force/moment sensor for the characteristic test of force control was manufactured, and the characteristic test for grasping an unknown object was performed using it. The fabricated gripper could grasp an unknown object stably. Thus, the developed 6-axis robot's finger force/moment sensor may be used for robot's gripper.