• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.8
• /
• pp.977-989
• /
• 1999

The purpose of this study is to analyze how the human body having more muscles than its degree-of-freedom modulates an effective stiffness using redundant actuation, and to apply this concept to the design and control of advanced machines which requires adaptable spring. To investigate the adaptable stiffness phenomenon due to redundant actuation in the human body, this paper derives a general stiffness model of the Human body. In particular, for a planar 1 DOF human arm model, a planar 2 DOF human arm model, a spherical 3 DOF shoulder model, a 4 DOF human arm model, and a 7 DOF human arm model, the required nonlinear geometry ad the number of required actuator for successful modulation of the effective stiffness are analyzed along with a load distribution method for modulation of the required stiffness of such systems. Secondly, the concept of motion frequency modulation is introduced to show the usefulness of adaptive stiffness modulation. The motion frequency modulation represents a control of stiffness and / or inertia properties of systems. To show the effectiveness of the proposed algorithm, simulations are performed for 2 DOF anthropomorphic robot.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.8 no.2
• /
• pp.108-115
• /
• 1999

This paper proposes the robot inverse calibration method using a neural networks. A highorder networks called Pi-Sigma networks has been used. The Pi-Sigma networks uses linear summing units in the hidden layer and product unit in output layer. The inverse calibration model which compensates the difference of joint variables only between measuring value and analytic value about the desired pose(position orientation) of a robot is proposed. The compensated values are determined by using the weights obtained from the learning process of the neural networks previously. To prove the reasonableness the SCARA type direct drive robot(4-DOF) and anthropomorphic robot(6-DOF) are simulated. It shows that the proposed calibration method can reduce the errors of the joint variables from $\pm$3 to $\pm$0.1.

• Journal of Institute of Control, Robotics and Systems
• /
• v.14 no.1
• /
• pp.62-69
• /
• 2008

A specialized anthropomorphic robot hand which can be attached to the biped humanoid robot MAHRU-R in KIST, has been developed. This built-in type hand consists of three fingers and a thumb with total four DOF(Degrees of Freedom) where the finger mechanism is well designed for grasping typical objects stably in human's daily activities such as sphere and cylinder shaped objects. The restriction of possible motions and the limitation of grasping objects arising from the reduction of DOF can be overcome by reflecting a typical human finger's motion profile to the design procedure. As a result, the developed hand can imitate not only human hand's shape but also its motion in a compact and efficient manner. Also this novel robot hand can perform various human hand gestures naturally and grasp normal objects with both power and precision grasping capability.

• Journal of the Korean Society for Precision Engineering
• /
• v.14 no.12
• /
• pp.86-94
• /
• 1997

This paper proposes the robot inverse calibration method using a neural networks. A high-order networks called Pi-Sigma networks has been used. The Pi-Sigma networks uses linear summing units in the hidden layer and product unit in output layer. The inverse calibration model which compensates the diff- erence of joint variables only between measuring value and analytic value about the desired pose(position, orientation) of a robot is proposed. The compensated values are determined by using the weights obtained from the learning process of the neural networks previously. To prove the reasonableness, the SCARA type direct drive robot(4-DOF) and anthropomorphic robot(6-DOF) are simulated. It shows that the proposed calibration method can reduce the errors of the joint variables from .+-. 5 .deg. to .+-. 0.1 .deg. .

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.9
• /
• pp.363-370
• /
• 2022

Grasping a target object among clutter objects without collision requires machine intelligence. Machine intelligence includes environment recognition, target & obstacle recognition, collision-free path planning, and object grasping intelligence of robot hands. In this work, we implement such system in simulation and hardware to grasp a target object without collision. We use a RGB-D image sensor to recognize the environment and objects. Various path-finding algorithms been implemented and tested to find collision-free paths. Finally for an anthropomorphic robot hand, object grasping intelligence is learned through deep reinforcement learning. In our simulation environment, grasping a target out of five clutter objects, showed an average success rate of 78.8%and a collision rate of 34% without path planning. Whereas our system combined with path planning showed an average success rate of 94% and an average collision rate of 20%. In our hardware environment grasping a target out of three clutter objects showed an average success rate of 30% and a collision rate of 97% without path planning whereas our system combined with path planning showed an average success rate of 90% and an average collision rate of 23%. Our results show that grasping a target object in clutter is feasible with vision intelligence, path planning, and deep RL.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.1232-1235
• /
• 2005

In recent years, as the robot technology is developed the researches on the artificial muscle actuator that enable robot to move dextrously like biological organ become active. The widely used materials for artificial muscle are the shape memory alloy and the electroactive polymer. These actuators have the higher energy density than the electromechanical actuator such as motor. However, there are some drawbacks for actuator. SMA has the hysterical dynamic characteristics. In this paper the segmented binary control for reducing the hysteresis of SMA is proposed and the simulation of anthropomorphic robotic hand is performed using ADAMS.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2005.05a
• /
• pp.144-149
• /
• 2005

In recent years, as the robot technology is developed the researches on the artificial muscle actuator that enable robot to move dextrously like biological organ become active. The widely used materials for artificial muscle are the shape memory alloy and the electroactive polymer. These actuators have the higher energy density than the electromechanical actuator such as motor. However, there are some drawbacks for actuator. SMA has the hysterical dynamic characteristics. In this paper the segmented binary control for reducing the hysteresis of SMA is proposed and the simulation of anthropomorphic robotic hand is performed using ADAMS.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.15 no.5
• /
• pp.14-20
• /
• 2006

In recent year, as the robot technology is developed, the researches on the artificial muscle actuator that enables robot to move dexterously like biological organ become active. Actuators are key technologies underpinning robotics. Breakthroughs in actuator technology, particular in terms of power-to-weight ratio, or energy-density, will have significant impacts upon the design and control of robotic system. In this paper, a new approach to design and control of shape memory alloy(SMA) actuator is presented to drive the robot hand. SMA wire is divided into many segments and their thermal states of the SMA are controlled individually in a binary manner. This control manner will reduce the hysteresis that the SMA material has and it becomes the fundamental technology to develop the anthropomorphic robot hand. In this paper, the mechanism In the digital step motor of the shape memory alloy that is driven by the segmented binary control, which is a new control technique, is studied. This SMA digital step actuator applies for the robot hand and the driving mechanism of the robot hand is investigated.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.8
• /
• pp.656-664
• /
• 2016

While anthropomorphic robots are gaining interest, dual-arm robots are widely used in the industrial environment. Many methods exist in order to implement bimanual tasks by dual-arm robot. However, kinesthetic teaching is used in this paper. This paper suggests three different kinesthetic teaching methods that can implement most of the human task by the robot. The three kinesthetic teaching methods are joint level, task level, and contact level teaching. The task introduced in this paper is box packing, which is a popular and complex task in industrial environment. The task is programmed into the dual-arm robot by utilizing the suggested kinesthetic teaching method, and this paper claims that most tasks can be implemented by using the suggesting kinesthetic teaching methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.78.6-78
• /
• 2002

$\textbullet$ The Gifu Hand III is a 5-fingered hand driven by built-in servomotors and has 20 joints with 10 DOF. $\textbullet$ The backlash of transmission, the mobility space, and the opposability of the thumb are improved. $\textbullet$ The new distributed tactile sensor with 859 detecting points is mounted on the hand surface. $\textbullet$ Experiments of grasping objects by a grasping strategy imitating human grasping reflex are shown.